MCP server de Sequenzy
Commit 13 nov 2026Conecta Sequenzy con clientes de IA para gestionar operaciones de email marketing, secuencias y campañas mediante herramientas estructuradas.
También conocido como Sequenzy/mcp, mcp, @sequenzy/mcp
- Stars
- 3
- Herramientas
- 226
- Descargas, 30 días
- —
- Última versión
- 0.0.255
- Último commit
- 13 nov 2026
- Licencia
- MIT
¿Qué es Sequenzy MCP?
Conecta Sequenzy con clientes de IA para gestionar operaciones de email marketing, secuencias y campañas mediante herramientas estructuradas. Expone 226 herramientas, como open_sequenzy, preview_sequence, open_image_upload.
Resumen generado a partir del README del proyecto
Acerca del MCP server de Sequenzy
Sequenzy MCP conecta la plataforma de email marketing Sequenzy con asistentes de IA como Claude Desktop, Cursor y Windsurf. Permite a desarrolladores y especialistas en marketing gestionar operaciones de correo electrónico, secuencias y campañas utilizando herramientas estructuradas en lugar de llamadas manuales a la API. El servidor permite a los usuarios realizar tareas complejas como redactar campañas para hasta 100,000 contactos, gestionar etiquetas de suscriptores y sincronizar segmentos con audiencias personalizadas de Meta.
Los usuarios pueden interactuar con sus flujos de trabajo de marketing mediante vistas previas visuales que renderizan correos para vistas de escritorio y móviles. El servidor admite una gestión avanzada de secuencias, lo que incluye la visualización de disparadores, retrasos y ramificaciones a través de un diagrama conectado. La configuración se simplifica mediante un asistente de línea de comandos que gestiona el inicio de sesión en el navegador, la creación de la clave API y la configuración del cliente. Aunque el servidor proporciona widgets visuales enriquecidos para clientes compatibles, también devuelve HTML estructurado y datos para clientes que no cuentan con soporte de aplicaciones especializadas.
Qué puedes hacer con él
- 1Previsualización de renderizado de correosMuestra cómo se verá una plantilla de correo específica tanto en dispositivos de escritorio como móviles.
- 2Visualización de flujos de automatizaciónExamina la lógica de una secuencia de marketing, incluyendo sus ramificaciones y retrasos.
- 3Gestión de grandes grupos de contactosCrea un nuevo borrador de campaña para un segmento de audiencia específico.
- 4Monitoreo de importaciones de datosRealiza el seguimiento del progreso y la tasa de éxito de una tarea de importación de suscriptores.
Escrito a partir del README del proyecto. Cada caso de uso se verifica con una cita del mismo.
Cómo instalar Sequenzy MCP
Formatos verificados el 10 nov 2026Añadir a CursorAñadir a GooseInstalar Sequenzy MCP en Claude Code
$ claude mcp add --env 'SEQUENZY_API_KEY=<YOUR_SEQUENZY_API_KEY>' mcp -- npx -y @sequenzy/mcpInstalar Sequenzy MCP en Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@sequenzy/mcp"
],
"env": {
"SEQUENZY_API_KEY": "<YOUR_SEQUENZY_API_KEY>"
}
}
}
}Instalar Sequenzy MCP en VS Code
.vscode/mcp.json
{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@sequenzy/mcp"
],
"env": {
"SEQUENZY_API_KEY": "<YOUR_SEQUENZY_API_KEY>"
}
}
}
}Instalar Sequenzy MCP en VS Code CLI
$ code --add-mcp '{"name":"mcp","type":"stdio","command":"npx","args":["-y","@sequenzy/mcp"],"env":{"SEQUENZY_API_KEY":"<YOUR_SEQUENZY_API_KEY>"}}'Instalar Sequenzy MCP en Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@sequenzy/mcp"
],
"env": {
"SEQUENZY_API_KEY": "<YOUR_SEQUENZY_API_KEY>"
}
}
}
}Instalar Sequenzy MCP en Codex
$ codex mcp add mcp --env 'SEQUENZY_API_KEY=<YOUR_SEQUENZY_API_KEY>' -- npx -y @sequenzy/mcpInstalar Sequenzy MCP en Codex config.toml
~/.codex/config.toml
[mcp_servers.mcp] command = "npx" args = ["-y", "@sequenzy/mcp"] [mcp_servers.mcp.env] SEQUENZY_API_KEY = "<YOUR_SEQUENZY_API_KEY>"
Instalar Sequenzy MCP en Gemini CLI
$ gemini mcp add -s user -e 'SEQUENZY_API_KEY=<YOUR_SEQUENZY_API_KEY>' mcp npx -- -y @sequenzy/mcpInstalar Sequenzy MCP en Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@sequenzy/mcp"
],
"env": {
"SEQUENZY_API_KEY": "<YOUR_SEQUENZY_API_KEY>"
}
}
}
}Iniciado por HostedMCP el 16 nov 2026
Ejecutamos este comando para @sequenzy/mcp 0.0.255 en un contenedor aislado sin red ni claves. Respondió y listó 226 herramientas.
Herramientas de Sequenzy MCP 226
Leído del servidor con tools/listopen_sequenzySolo lecturaOpen the Sequenzy workspace view: a company's sequences and recent campaigns, each opening as a flow diagram or rendered email in MCP Apps clients. Read-only. Without companyId it uses the selected company and offers a switcher. The open view has its own tools (get_view_state, reload_view and others); call reload_view after editing what the user is looking at. Other clients receive the lists as data.
preview_sequenceSolo lecturaShow an interactive visual preview of a saved sequence in clients that support MCP Apps. Displays its real nodes, connections, delays and branches; click an email step to preview its saved rendered email, including available A/B variants, at desktop or mobile width. Read-only: does not activate, enroll or send. Requires an explicit companyId to keep step previews pinned to the same company. Other clients receive the sequence data; use render_email for individual step HTML.
open_image_uploadSolo lecturaOpen a compact image picker as a supporting workflow for email creation. Opening does not upload anything; the user chooses an image and explicitly submits it through upload_image_asset. Requires an explicit companyId. Clients without MCP Apps can use upload_image_asset directly.
monitor_subscriber_importSolo lecturaShow a compact progress card for an existing subscriber import. Reads immediately, then polls get_subscriber_import every five seconds for up to five minutes while running. Refresh resumes monitoring. Requires explicit companyId and importId. Does not start or retry an import. Other clients receive the import data.
get_accountSolo lecturaGet your own Sequenzy account information (not your customers' B2B accounts; use list_accounts for those) including available companies, the current API key's permission scopes, and the API Keys management URL. Useful after a missing-scope error to inspect access and find the replacement-key settings. When the key has access to multiple companies, other tools need a target company: either one selected with select_company or a companyId passed in each call. The response shows 'companies' (all available) and 'selectedCompanyId' (currently active). Subsequent operations use the selected company unless a companyId override is passed.
select_companyAcciónSelect which company to operate on (for user-scoped API keys with access to multiple companies). Use get_account to see available companies. After selecting, all subsequent operations will use this company unless you pass a companyId override.
get_app_urlsSolo lecturaGenerate Sequenzy dashboard URLs for known resource IDs. Use this when the user asks where to review or edit a generated sequence, campaign, template, company settings, or subscription. The response always includes the canonical owner-facing subscription URL when a company is resolved. For compatibility, settingsTab billing or subscription also resolves to that Account page rather than a nonexistent Settings tab. If companyId is omitted, the selected/current company is used when available.
create_companyAcciónCreate a company with a website domain, or set noWebsite and provide a name. Optionally provide a business description and create a four-email welcome sequence. Requires a personal account key (seq_user_...); company-scoped keys cannot create another workspace. Website creation queues brand processing and this tool polls for completion.
get_companySolo lecturaRead one company's settings and status: processing status, product info, brand colors, AI writing context, emailStylePreference (visual, plain, or null when not chosen), websiteType (saas, content, ecommerce, offline, mobile_app, other or null) with websiteTypeSource (user when chosen, detected from the website), header and footer (emailPreset), footerStyle for new AI-designed emails (auto, classic, mono, band or editorial), footerTheme (the auto-added footer's resolved colors and copy; change it with update_company footerTheme), reply-tracking settings, effective email localization settings, and darkModeLogoStatus (pending, outlined, not_needed or custom) with darkModeLogoUrl, the logo dark-mode inboxes show (set your own with update_company darkModeLogoUrl). The read-only emailBranding says whether "Sent with Sequenzy" is visible, whether the owner subscription removes it, why, the tier/status, the upgrade or renewal action, and subscriptionUrl. Branding is added at send time, so a paid entitlement removes it from existing live sequences without editing their emails. defaultSubscriberListIds are the lists new contacts join when nothing targets them: JSON null means every current and future list, not an empty selection; [] means no list; an array means exactly those lists. Change it with update_company. Read it before connecting an integration without per-integration list targeting (Dodo Payments, PostHog, Polar, Paddle, and similar): its live contacts and payment-provider backfills land there. PostHog history imports are different: their contacts get no list memberships. Send Time Optimization is not a company or sequence setting. It is campaign-only: audit with list_campaigns or get_campaign (sendTimeOptimization, sendTimeWindowHours, default 12h window). Change a draft with update_campaign, or apply it when calling schedule_campaign. Sequences use sendingWindow (allowed local hours/days), which is a shared gate, not per-subscriber predicted send times.
update_companyAcciónEdit company product info, brand context, the default email theme, reply-tracking settings, account-wide sending identity defaults, and defaultSubscriberListIds - which lists new untargeted contacts join. That field is the write path for Dodo Payments and live PostHog events, plus every other integration without per-integration targeting: null = every current and future list (not 'no lists'), [] = no list, [ids] = exactly those lists. PostHog history imports always create contacts without list memberships. fromEmail must use a verified sending domain; replyTo may be any valid mailbox. Missing sender/reply profiles are created automatically. Send fromName/replyToName on their own to rename the display name of the existing default profile without changing its address, or pair them with senderProfileId/replyProfileId (from list_sender_profiles) to rename a specific profile. Provide at least one editable field. Profile, branding, and AI-context fields need the company_profile:manage scope; the sending identity, reply-tracking, and defaultSubscriberListIds fields additionally need companies:manage.
get_shopify_automation_settingsSolo lecturaGet the connected Shopify store's automation settings: browse abandonment (emails shoppers who viewed a product but didn't buy), cart abandonment (fires ecommerce.cart_abandoned with the full cart after the cart sits inactive), and price drop alerts (emails recent viewers when a product's price falls). Returns the effective values with platform defaults applied.
update_shopify_automation_settingsAcciónUpdate the connected Shopify store's browse-abandonment, cart-abandonment, and/or price-drop automation settings. Partial update: omitted sections are untouched, omitted fields within a section keep their current value, and passing null for a section resets it to the platform defaults (browse abandonment: on, 2h delay, 24h cooldown; cart abandonment: on, 1h inactivity, 24h cooldown; price drop: on, 5% minimum drop, 30-day viewer lookback, 7-day cooldown).
| open_sequenzy | Open the Sequenzy workspace view: a company's sequences and recent campaigns, each opening as a flow diagram or rendered email in MCP Apps clients. Read-only. Without companyId it uses the selected company and offers a switcher. The open view has its own tools (get_view_state, reload_view and others); call reload_view after editing what the user is looking at. Other clients receive the lists as data. | Solo lectura |
| preview_sequence | Show an interactive visual preview of a saved sequence in clients that support MCP Apps. Displays its real nodes, connections, delays and branches; click an email step to preview its saved rendered email, including available A/B variants, at desktop or mobile width. Read-only: does not activate, enroll or send. Requires an explicit companyId to keep step previews pinned to the same company. Other clients receive the sequence data; use render_email for individual step HTML. | Solo lectura |
| open_image_upload | Open a compact image picker as a supporting workflow for email creation. Opening does not upload anything; the user chooses an image and explicitly submits it through upload_image_asset. Requires an explicit companyId. Clients without MCP Apps can use upload_image_asset directly. | Solo lectura |
| monitor_subscriber_import | Show a compact progress card for an existing subscriber import. Reads immediately, then polls get_subscriber_import every five seconds for up to five minutes while running. Refresh resumes monitoring. Requires explicit companyId and importId. Does not start or retry an import. Other clients receive the import data. | Solo lectura |
| get_account | Get your own Sequenzy account information (not your customers' B2B accounts; use list_accounts for those) including available companies, the current API key's permission scopes, and the API Keys management URL. Useful after a missing-scope error to inspect access and find the replacement-key settings. When the key has access to multiple companies, other tools need a target company: either one selected with select_company or a companyId passed in each call. The response shows 'companies' (all available) and 'selectedCompanyId' (currently active). Subsequent operations use the selected company unless a companyId override is passed. | Solo lectura |
| select_company | Select which company to operate on (for user-scoped API keys with access to multiple companies). Use get_account to see available companies. After selecting, all subsequent operations will use this company unless you pass a companyId override. | Acción |
| get_app_urls | Generate Sequenzy dashboard URLs for known resource IDs. Use this when the user asks where to review or edit a generated sequence, campaign, template, company settings, or subscription. The response always includes the canonical owner-facing subscription URL when a company is resolved. For compatibility, settingsTab billing or subscription also resolves to that Account page rather than a nonexistent Settings tab. If companyId is omitted, the selected/current company is used when available. | Solo lectura |
| create_company | Create a company with a website domain, or set noWebsite and provide a name. Optionally provide a business description and create a four-email welcome sequence. Requires a personal account key (seq_user_...); company-scoped keys cannot create another workspace. Website creation queues brand processing and this tool polls for completion. | Acción |
| get_company | Read one company's settings and status: processing status, product info, brand colors, AI writing context, emailStylePreference (visual, plain, or null when not chosen), websiteType (saas, content, ecommerce, offline, mobile_app, other or null) with websiteTypeSource (user when chosen, detected from the website), header and footer (emailPreset), footerStyle for new AI-designed emails (auto, classic, mono, band or editorial), footerTheme (the auto-added footer's resolved colors and copy; change it with update_company footerTheme), reply-tracking settings, effective email localization settings, and darkModeLogoStatus (pending, outlined, not_needed or custom) with darkModeLogoUrl, the logo dark-mode inboxes show (set your own with update_company darkModeLogoUrl). The read-only emailBranding says whether "Sent with Sequenzy" is visible, whether the owner subscription removes it, why, the tier/status, the upgrade or renewal action, and subscriptionUrl. Branding is added at send time, so a paid entitlement removes it from existing live sequences without editing their emails. defaultSubscriberListIds are the lists new contacts join when nothing targets them: JSON null means every current and future list, not an empty selection; [] means no list; an array means exactly those lists. Change it with update_company. Read it before connecting an integration without per-integration list targeting (Dodo Payments, PostHog, Polar, Paddle, and similar): its live contacts and payment-provider backfills land there. PostHog history imports are different: their contacts get no list memberships. Send Time Optimization is not a company or sequence setting. It is campaign-only: audit with list_campaigns or get_campaign (sendTimeOptimization, sendTimeWindowHours, default 12h window). Change a draft with update_campaign, or apply it when calling schedule_campaign. Sequences use sendingWindow (allowed local hours/days), which is a shared gate, not per-subscriber predicted send times. | Solo lectura |
| update_company | Edit company product info, brand context, the default email theme, reply-tracking settings, account-wide sending identity defaults, and defaultSubscriberListIds - which lists new untargeted contacts join. That field is the write path for Dodo Payments and live PostHog events, plus every other integration without per-integration targeting: null = every current and future list (not 'no lists'), [] = no list, [ids] = exactly those lists. PostHog history imports always create contacts without list memberships. fromEmail must use a verified sending domain; replyTo may be any valid mailbox. Missing sender/reply profiles are created automatically. Send fromName/replyToName on their own to rename the display name of the existing default profile without changing its address, or pair them with senderProfileId/replyProfileId (from list_sender_profiles) to rename a specific profile. Provide at least one editable field. Profile, branding, and AI-context fields need the company_profile:manage scope; the sending identity, reply-tracking, and defaultSubscriberListIds fields additionally need companies:manage. | Acción |
| get_shopify_automation_settings | Get the connected Shopify store's automation settings: browse abandonment (emails shoppers who viewed a product but didn't buy), cart abandonment (fires ecommerce.cart_abandoned with the full cart after the cart sits inactive), and price drop alerts (emails recent viewers when a product's price falls). Returns the effective values with platform defaults applied. | Solo lectura |
| update_shopify_automation_settings | Update the connected Shopify store's browse-abandonment, cart-abandonment, and/or price-drop automation settings. Partial update: omitted sections are untouched, omitted fields within a section keep their current value, and passing null for a section resets it to the platform defaults (browse abandonment: on, 2h delay, 24h cooldown; cart abandonment: on, 1h inactivity, 24h cooldown; price drop: on, 5% minimum drop, 30-day viewer lookback, 7-day cooldown). | Acción |
Mostrar las 226 herramientasMostrar menos herramientas
get_sync_rulesSolo lecturaGet the company's sync rules: the automatic tag changes applied when events fire (e.g. order placed -> add a tag). New companies start with the payment and ecommerce lifecycle rules; some older ones have none. Returns the effective rules plus isDefault, which is true when the company has explicitly opted into the inherited SaaS/ecommerce platform preset.
update_sync_rulesAcciónReplace the company's sync rules. Pass the FULL rule set (fetch with get_sync_rules first and edit it - this is not a partial update), [] to disable rules, or null to opt into the inherited SaaS/ecommerce platform preset. Each rule: { triggerEvent, actions: { addTags, removeTags }, conditions? }. Conditions support requiresTags, requiresNotTags, and purchasedProduct ({ tags?, collectionIds?, productTypes?, vendors? }) which matches products on commerce events - e.g. tag buyers of products carrying a given product tag.
get_email_design_systemSolo lecturaGet the company's email design system: the visual identity every AI-generated email (campaigns and sequence steps) renders inside - design code (kicker style, title alignment, button shape, divider style, density, sanctioned opener treatments) plus the composition spine (hero-led | editorial | product-spec) that anchors email structure. The identity is stored as the company's emailDesignPrompt direction text; this tool returns the structured tokens parsed from that text (falling back to brand derivation per unstated token), the raw directionText, and isDefault (true while the identity is still purely derived from brand context).
update_email_design_systemAcciónAdjust the company's email design system (partial update - only pass the fields to change). Affects every future AI-generated email and sequence enrichment. designCode tokens: kickerStyle (chip | letterspaced | none), titleAlignment (left | center), buttonShape (rounded | pill), dividerStyle (line | space), density (airy | compact), openerTreatments (subset of hero-image, hero-color, header-hero, editorial-masthead, title-led). compositionSpine: hero-led | editorial | product-spec. The adjustment is written into the company's emailDesignPrompt direction text (the single source of truth): the new identity's sentences are prepended and any custom prose the text carried is preserved below them. Pass reset: true to clear the direction text and return to brand-derived defaults. Free-form prose edits belong on update_company's emailDesignPrompt instead.
get_design_mdSolo lecturaGet the company's design.md: its email design brief as one Markdown document. YAML frontmatter holds the exact values Sequenzy applies when rendering AI-generated emails (logo, colors, fonts.body/heading, buttons.style/radius, cornerRadius, layout tokens); prose sections hold the Voice and the Design direction the AI writer follows, plus a read-only Reference email naming the saved AI style. Values shown are the ones emails actually get. Returns markdown, version (pass it to update_design_md as expectedVersion), and the parsed designSystem. Read this before writing or redesigning emails for the company.
update_design_mdAcciónSave an edited design.md (full document from get_design_md). Only values you changed are written: a frontmatter value left as returned is not written and keeps its current behavior, a changed radius or heading font is pinned, and a changed layout token is written into the design direction ahead of its prose. Removing a key or section leaves it unchanged; an empty Voice or Design direction section clears it (an empty direction returns to brand-derived defaults). The Reference email section is read-only. Invalid values fail with line-numbered issues; unknown keys are ignored with warnings. Pass dryRun: true to preview changes and warnings without saving. Pass expectedVersion from get_design_md; on a 409 conflict, read again and reapply your edits.
create_api_keyAcciónCreate a new API key for a company, typically when setting up a Sequenzy integration in a project. The CLI reads the key from the SEQUENZY_API_KEY environment variable; SDK clients accept it through the apiKey option, such as apiKey: process.env.SEQUENZY_API_KEY. The key is returned only once at creation time and cannot be retrieved later.
request_api_key_handoffSolo lecturaPrepare an owner-confirmed handoff link for creating or rotating an API key. Intended for when key management fails because the active key is missing api_keys:manage; that scope fails closed on purpose, and this handoff is the supported path. Returns a URL that opens the dashboard's create-key form with the requested name and permissions already filled in. This tool creates nothing and never returns a key: the workspace owner opens the URL, reviews the form, clicks Create, and copies the new key from the browser. To rotate, pass replaceApiKeyId (or the literal string 'current' for the key you are authenticated with) and the dashboard offers to revoke the predecessor after the replacement is created. Requires only account:read.
list_api_keysSolo lecturaList company-scoped API keys as non-secret metadata. Returns IDs, names, prefixes, permissions, usage timestamps, and whether each key is the active credential. It never returns a plain key or stored key hash. Use this before revoke_api_key or delete_api_key to identify the exact unused key.
update_api_keyAcciónRename a company-scoped API key or replace its permissions in place, without issuing a new key. Use this when a call fails with a missing-scope error: the key value stays the same, so no client has to be re-wired, and added permissions apply on the next retry. Removed permissions may remain usable for up to five minutes while API caches expire. preset and scopes REPLACE the current selection rather than merging into it - call list_api_keys first and pass the full set you want. Returns non-secret metadata only.
revoke_api_keyAcciónPermanently revoke a company-scoped API key by ID. The response contains non-secret metadata only. Call list_api_keys first and compare the ID, name, prefix, and isCurrent flag because revoking a key cannot be undone and may invalidate the active credential.
delete_api_keyAcciónCompatibility alias for revoke_api_key. Permanently delete a company-scoped API key by ID and return non-secret metadata only. Call list_api_keys first to verify the exact target.
list_websitesSolo lecturaList configured sending domains with separate DNS verification, sending readiness, and SPF, DKIM, and MAIL FROM status.
add_sending_domainAcciónAdd and configure a sending domain. A domain belongs to exactly one company and cannot be moved or removed through the API, CLI or MCP afterwards, so confirm the target company with the user before adding. Before calling verify_sending_domain, publish the DKIM record from website.dnsRecords (when dnsRecords.platformDnsRecords is present, publish those CNAMEs instead: active and spare DKIM selectors plus a recommended bounces CNAME; only required entries block verification), the bounce records described by website.bounceRecords (its cnameRecord when style is cname, otherwise spf.record and mailFrom.mxRecord; never both at the same name), and DMARC when present. Tracked links use the company tracking domain shared by every sending domain; the first domain creates it under itself (links.<domain>) and returns its CNAME as dnsRecords.trackingRecord. It is optional and never blocks verification or sending: until it verifies, links use the shared Sequenzy domain. Manage it with set_tracking_domain.
add_websiteAcciónCompatibility alias for add_sending_domain. Adds a sending domain and returns the DNS records required for setup. The domain cannot be moved to another company afterwards; confirm the target company first.
check_websiteSolo lecturaRead a sending domain's separate DNS verification and sending readiness. Use verify_sending_domain to run a fresh check. dnsUpgrade.status: available (see upgrade_sending_domain_dns), in_progress (records to add), completed (oldRecords to delete) or unavailable. Receiving mail (inbox addresses and replies) is set up on receiving domains instead: see list_receiving_domains.
verify_sending_domainAcciónRun a fresh DNS check for a configured sending domain and return DNS verification separately from sending readiness. When readyToSend is false, read readiness.reason. Receiving mail (inbox addresses and replies) is checked on receiving domains instead: see verify_receiving_domain.
upgrade_sending_domain_dnsAcciónRecommended, optional one-time upgrade of a verified sending domain to the current per-domain CNAME DNS records, for better deliverability and security, without removing it; it keeps sending on current records and sending is normally not interrupted at the switch. Once started it runs to the switch. Needs website.dnsUpgrade.status available; an older SES-only domain also needs a DMARC record (400 otherwise). Publish the returned dnsUpgrade.records CNAMEs next to the current records, removing nothing; the domain switches by itself about an hour after the required ones are public (verify_sending_domain checks now, no skip). Removing dnsUpgrade.oldRecords (listed beforehand as replacedRecords) after deleteOldRecordsAfter (8 days) is optional; each is marked removed once gone from public DNS.
configure_sending_domainAcciónDeprecated: tracking is company-wide; prefer set_tracking_domain. Sets <trackingPrefix>.<domain> as the company tracking domain when the company has none. Repeating the company's current tracking hostname is a no-op; a different one changes nothing and the response message says so (use set_tracking_domain to change it). Publish the returned website.tracking.cnameRecord; sending never waits for it.
list_integrationsSolo lecturaList connected integrations (Stripe, Shopify, Supabase, Clerk, WooCommerce, ad platforms, and so on) with connection status, sync state, last sync time, and last sync error. Read-only: credentials, access tokens, and webhook secrets are never returned. Use this to audit which external systems feed this account and whether their syncs are healthy.
list_sender_profilesSolo lecturaList sender (From) profiles and reply-to profiles for the company, including which are the account defaults and whether each sender address has verified DNS plus a ready home transport. Use this to audit sending identity before scheduling, or to pick a valid senderProfileId/replyProfileId for create_campaign or update_campaign.
update_sender_profileAcciónRename one sender (From) or reply-to profile in place, without changing which profile is the account default. Use this to standardize a display name across the several identities a mailbox may carry, e.g. renaming 'Viraj from SnapCount' to 'SnapCount'. Get IDs from list_sender_profiles. Only the display name changes: the address, its sending domain, and the account-wide default From/Reply-To selection are untouched, and existing campaigns and sequences pinned to this profile pick up the new name. To change which profile is the default instead, use update_company with senderProfileId/replyProfileId. A company cannot have two sender identities with the same name on one address, so a colliding rename is rejected and reports the ID already using that name. Requires the companies:manage scope.
delete_sender_profileAcciónPermanently delete one sender (From) profile. This cannot be undone. Deletion is refused when this is the company's last sender or when a live campaign, active sequence (including a step-level override), or transactional email still uses it. Draft and rejected campaigns plus the account default are reassigned to the best remaining sender when needed; read fallbackSenderProfileId in the result and review that identity before sending. Reply-to profiles are not supported by this tool. Get the profile ID from list_sender_profiles. Requires the companies:manage scope.
get_sending_statusSolo lecturaCheck whether company-level email sending is active, paused, or suspended, and what it takes to restore it. Relevant whenever a send, test send, or sequence step fails for a reason that is not a validation error, and whenever the account reports a bounce or complaint rate problem. Returns the pause reason and reason kind, when it was paused, the automated sender-health review state, whether the one-click resume is currently available (and if not, which gate is blocking it), the enforcement counts and thresholds for permanent bounces, temporary bounces, and complaints, and ordered remediation steps. There is no rolling window and no expiry. Enforcement uses all-time totals counted from a reset watermark, so a paused rate does NOT decay over time. Waiting does not restore sending. The rate only changes as real (non-test) sends accumulate or when a resume moves the watermark. `metricsWindow` contains the exact watermark timestamps. Requires the account:read scope, so a read-only key can diagnose a blocked account. resume_sending restores sending once the underlying problem is fixed.
resume_sendingAcciónRequest that paused company-level sending be restored after fixing the cause. This is the supported remediation path for an account paused by the permanent-bounce rate limit - it is not a bypass and it never removes suppressions. Intended for after the bad source (import, form, or integration) is actually fixed while permanent bounces remain suppressed. get_sending_status reports the current gate: for a paused workspace, resume only restores sending when `selfResume.canSelfResume` is true, which requires a hard-bounce pause reason, a cleared automated sender-health review, and no admin block. A workspace that is already active succeeds as a no-op with `resumed: false`; any other blocked state needs a support review, and the error names which one. listSanitizationConfirmed: true is required and represents the user's confirmation that the remediation is done; it is recorded on the account's audit trail as their statement that the list is clean. On success, sending is restored, the bounce watermark resets so the rate is recalculated from later sends, and the service attempts to requeue paused campaigns plus due sequence steps. The response message contains support guidance if part of that queue handoff is temporarily unavailable. Requires the companies:manage scope and owner or admin access to the company.
get_tracking_domainSolo lecturaGet the company tracking domain every sending domain uses for tracked links and opens: hostname, status, whether links use it now (active), the CNAME record to publish, and the last check. trackingDomain is null when none is set and links use the shared Sequenzy domain.
set_tracking_domainAcciónSet or change the company tracking domain: one hostname every sending domain uses for tracked links and opens, for example links.example.com. Any subdomain the user controls works; a bare root domain is rejected. Publish the returned trackingDomain.cnameRecord. It never blocks sending: until it verifies, links use the shared Sequenzy domain. Changing it keeps the previous hostname serving links in emails already sent while its old CNAME stays. Confirm the hostname with the user first.
verify_tracking_domainAcciónCheck the company tracking domain's CNAME and HTTPS certificate now instead of waiting for the automatic check. Returns verified, a message with the next step, and the tracking domain status.
remove_tracking_domainAcciónRemove the company tracking domain. New emails use the shared Sequenzy tracking domain, and links and images in emails already sent through the removed domain stop working. Confirm with the user first.
get_frequency_capSolo lecturaGet the company frequency cap: the most marketing emails (campaigns plus marketing sequence emails) one contact may receive in a rolling window. frequencyCap is null when no cap is set. Only sequences with frequencyCapEnabled respect it: their email steps are skipped for a contact already at the cap, and the contact moves to the next step. Campaigns count toward the cap but always send. Transactional emails never count and are never skipped.
set_frequency_capAcciónSet or change the company frequency cap: at most maxEmails marketing emails per contact in any windowHours. Applies from the next email step onward to sequences with frequencyCapEnabled (set it with create_sequence or update_sequence). Campaigns count toward the cap but always send. Transactional emails never count and are never skipped.
clear_frequency_capAcciónTurn the company frequency cap off. Sequences keep their frequencyCapEnabled setting, but nothing is skipped until a cap is set again with set_frequency_cap.
get_tracking_settingsSolo lecturaGet the company's email tracking and signup consent configuration: account-wide and Transactional API open/click defaults, unsubscribe tracking, opt-in strict bot filtering, default attribution window, automatic UTM tagging, the company tracking domain (shared by every sending domain) and its verification status, inbound reply-tracking settings, and whether double opt-in is required for new contacts. Use this to audit measurement readiness, to explain why opens or clicks may not be recorded, and to check how contacts enter the list before investigating bot or alias signups.
update_tracking_settingsAcciónUpdate the company's email tracking defaults: account-wide open and click tracking, narrower open and click defaults for Transactional API sends, unsubscribe tracking, opt-in strict bot filtering, the default attribution window, and automatic UTM tagging, plus the account-wide double opt-in requirement for new contacts. Applies to emails sent afterwards; already-sent emails keep the links they were rendered with. Transactional API defaults cannot enable an account-wide setting that is off, and an individual send may opt out again. Provide at least one field. Reply tracking (inbound email, reply domain mode, reply forwarding) is on update_company. The company tracking domain is managed with set_tracking_domain, verify_tracking_domain and remove_tracking_domain.
get_notification_preferencesSolo lecturaGet the account notification settings for the API key's own user in this company: whether Sequenzy emails them when a new subscriber joins, when a form is submitted, when a campaign finishes sending, and whether they get the Monday weekly report (last week's sends, engagement, new subscribers, revenue, goals, and sequence trends versus the week before; on by default for the workspace owner, off by default for invited members, only sent for weeks with more than 10 emails sent), whether they get the account health check's findings (account_health: on by default for the owner, opt-in for members; carried inside the Monday report when they get it, otherwise sent as their own email), and whether each arrives per-occurrence, as a daily summary, or weekly. Instant form-submission notifications stop after 50 per workspace per UTC day. Returns the supported modes and the platform defaults for this user's role alongside the current values. These are per-person settings - this never reads a teammate's preferences.
update_notification_preferencesAcciónChange which account notifications Sequenzy emails the API key's own user for this company. Requires the companies:manage scope. Useful before a bulk import or a large migration, when per-signup notifications would otherwise flood the inbox, or to turn the Monday weekly report off (it is on by default only for the workspace owner; invited members set weekly_report to 'weekly' to receive it), or to opt in or out of the account health check's findings (account_health 'instant' or 'off'). Modes are 'off', 'instant' (one email per occurrence), 'daily' (one summary per day), and 'weekly' (the Monday report); 'daily' is only valid for new_subscriber, because a campaign finishes once and each form submit is its own lead, and 'weekly' is only valid for weekly_report; account_health accepts only 'off' or 'instant'. Instant form-submission notifications stop after 50 per workspace per UTC day. New-subscriber notifications already fall back to a daily summary automatically on high-volume days, and imports never trigger them at all.
get_integration_guideSolo lecturaGet code examples for integrating Sequenzy into your project. SaaS and subscription apps: when the app uses Stripe, Polar, Paddle, Dodo, Lemon Squeezy, or Creem, the native integration in the Sequenzy dashboard (Settings -> Integrations) handles payment events and customer tags automatically, so no code is needed for those. E-commerce stores: Shopify and WooCommerce stores use the native integration in the dashboard, which syncs products, orders, and customers automatically. Other e-commerce stacks (custom checkouts, CheckoutChamp, Sticky.io, headless storefronts) integrate through the 'ecommerce' use case and the Commerce API. Manual event tracking code is an opt-in use case for custom events only. Use cases: - 'transactional': Sending one-off emails (welcome, password reset, receipts) - 'subscribe_form': Adding subscribers from signup forms - 'event_tracking': Tracking CUSTOM events only (not payment events - those come from the integration) - 'ecommerce': Connecting a custom e-commerce platform via the Commerce API (sync products, push orders/checkouts, power abandoned cart + back-in-stock automations) Protected server-side API work needs an API key (create_api_key), passed to the SDK as apiKey: process.env.SEQUENZY_API_KEY. Non-Webflow static-site saved forms are the exception: list_forms/create_form/get_form_embed return public embed code, and secret keys must never be placed in browser code. Webflow-connected companies use explicitly selected native forms in Settings -> Integrations -> Webflow; custom embed code is disabled.
get_account_healthSolo lecturaRead the company's account health report: what to fix in its email setup. Returns eligibility (checks need more than 1,000 active subscribers and an email sent in the last 14 days; hasMinimumSubscribers and sentRecently say which is missing), settings (enabled, autonomy), nextRunAt, the latest run (status, AI summary, stats snapshot of this period vs the one before) and runInProgress, plus open and snoozed insights, most severe first. Each insight has a detector (unhandled_event, stalled_sequence, campaign_gap, stale_campaign_drafts, underperforming_sequence_email, deliverability_spike, sending_domain_failing, missing_sequence: a standard lifecycle sequence such as welcome or re-engagement the workspace does not run yet, with the suggested playbook, name, goal, trigger and steps in evidence; over_emailing: people who got 4+ marketing emails within 24 hours in the last 7 days, with overloadedSubscribers, recipients, peakPerDay, frequencyCap and the campaigns, sequences and marketing Send API email (type api, which the cap does not limit) behind it as sources in evidence, where a sequence with enrollsEveryTime and peakEnrollmentsPerPerson 3 or more re-enrolls the same people and is fixed first with update_sequence enrollmentMode one_time; neglected_subscribers: active subscribers who got no marketing email in 90 days, with neglectedSubscribers, eligibleSubscribers and estimated in evidence; unengaged_signups: contacts who joined in the last 60 days with the same consumer mailbox provider (Gmail, Yahoo, Outlook; business domains are never reported) and first list, got 2+ emails each and almost never opened or clicked while other new signups do, usually a bot filling a signup form or a bought list, with provider, providerLabel, listId, listName, emailed, sends, engaged, baselineEmailed and baselineEngaged in evidence), severity, title, detail, recommendation fixAutomationId when the agent drafted an inactive sequence to fix it, and actions: what to do about it, primary first. Each action has kind (create_sequence, open_sequence, activate_sequence, pause_sequence, create_campaign, review_campaign_drafts, open_campaign, verify_sending_domain, open_domain_settings, review_deliverability, open_frequency_cap_settings, apply_frequency_cap, review_subscribers), label, targetId, a dashboard path for screens, a confirm prompt for changes to live sending, and the api call that performs it: create_sequence -> create_account_health_fix, activate_sequence -> enable_sequence, pause_sequence -> disable_sequence, verify_sending_domain -> verify_sending_domain, open_frequency_cap_settings -> set_frequency_cap (for example maxEmails 1, windowHours 16) when evidence.frequencyCap is null, otherwise get_frequency_cap and only tighten it (the cap only holds back sequences, so when the sources are only campaigns or api it does not help; suggest spacing those out instead), apply_frequency_cap -> update_sequence with frequencyCapEnabled true on targetId (only on sequences that can wait, such as tips and nudges, not welcome or onboarding); the others open dashboard screens. Each insight also has feedback: the API key owner's own answer from submit_account_health_feedback ({rating, comment, updatedAt}), or null. Confirm with the user before set_frequency_cap or apply_frequency_cap. Confirm with the user before activate_sequence or pause_sequence. Requires analytics:read.
update_account_health_settingsAcciónTurn the weekly account health check on or off and choose whether it drafts fixes. Pass only the fields to change. Checks never send email to subscribers or activate anything: autonomy draft_fixes only creates inactive draft sequences. Checks run every Monday at 05:00 UTC, before the Monday weekly report that carries the findings; the schedule is fixed. Who gets the findings is each person's own account_health notification preference (update_notification_preferences), not a workspace setting. Owner or admin only. Requires companies:manage.
run_account_health_checkAcciónQueue an account health check now and return the run. If a check is already queued or running it is returned with alreadyInProgress true instead. Manual checks are limited to one every 10 minutes and wait a day after a check that left findings open (both 429 RATE_LIMITED with retryAfterSeconds), need more than 1,000 active subscribers and an email sent in the last 14 days (409 NOT_ELIGIBLE), and never send the report email. The check usually finishes within a minute; poll get_account_health until runInProgress is false. Requires sequences:write.
update_account_health_insightAcciónAct on one account health insight by id (from get_account_health): dismiss hides it for good, snooze hides it for snoozeDays (default 7, up to 90), resolve marks it fixed (it reopens if the next check still finds it), reopen brings back a snoozed or resolved one. Requires sequences:write.
submit_account_health_feedbackAcciónTell the Sequenzy team whether an account health insight (id from get_account_health) was useful to the user: rating useful or not_useful, with an optional comment on why. Use it when the user says a finding or suggestion helped or missed, in their words; do not rate on their behalf. It records the API key owner's own answer and changes nothing in the workspace (to hide a finding use update_account_health_insight). One answer per person per insight: submitting again replaces the rating and keeps the earlier comment unless comment is passed (null clears it). Works for insights of any status. 404 for an unknown insight. Requires analytics:read.
create_account_health_fixAcciónBuild the draft sequence an account health insight asks for, from an unhandled_event or missing_sequence insight (its primary action kind is create_sequence). Creates an inactive sequence with its timing and empty, titled email steps: for unhandled_event, triggered by the event and named for the moment (for example New customer onboarding for saas.purchase); for missing_sequence, with the name, trigger (contact added, an event, or manual enrollment) and steps from the insight's evidence. Like a sequence from a template, a new draft has its emails written with AI: enrichmentStatus is processing while they are written (get_sequence reports progress), or not_queued when writing could not start (opening the sequence in the dashboard starts it; do not build again). Running it again returns the same draft with created false; steps that still have no email (for example a check's own draft nobody opened yet) are written then, with enrichmentStatus, and existing emails are never rewritten. 409 IN_PROGRESS when another build of the same insight is running; 400 for other insight types. Show it with get_sequence; enable_sequence only works once every email step has an email, and only after the user approves. Requires sequences:write.
get_email_ai_styleSolo lecturaGet the company's saved AI email appearance, revisionId and canManage. Read this before replacing or clearing a style. No active style returns null; an unsupported version can still have a revisionId for recovery.
save_email_ai_styleAcciónSave an email's appearance as the company default for future AI generation. Captures fonts, colors, spacing and block treatments, plus detected layout habits (for example dotted dividers around every button) and optional notes, without copying its content or modifying existing emails. Requires emails:write and access to the source email. Replaces only expectedStyleId; get and review the current style before replacing. All detected habits are kept unless layoutRuleIds narrows them; review style.layout.rules in the response and re-save with a subset if needed. Plain-text choices and explicit style requests take precedence during generation.
clear_email_ai_styleAcciónClear the company's saved AI appearance and restore normal defaults for future generations. Existing emails stay unchanged. Requires emails:write. Get and review the current style first; a conflict must not be retried automatically.
disconnect_integrationAcciónDisconnect a Lemon Squeezy or SparkLoop integration and stop live ingestion and sync. Deletes only the managed provider webhook; manual webhooks remain. Local disconnection succeeds even if provider cleanup fails: inspect cleanupWarning. Repeating the call retries Lemon Squeezy cleanup; for SparkLoop, delete the leftover webhook in SparkLoop. Other providers require the dashboard. Requires integrations:manage and owner/admin access.
connect_integrationAcciónConnect an API-key / webhook-secret integration: polar, paddle, dodo, lemon_squeezy, whop, revenuecat, superwall, adapty, creem, chargebee, clerk, posthog, segment, mixpanel, amplitude, affonso, attio, or sparkloop. Check list_integration_capabilities first - each connectable provider lists its exact connectFields there. Credentials are validated, stored encrypted, and never returned; payment providers queue their initial revenue backfill automatically. Lemon Squeezy creates a managed signed webhook when webhookSecret is omitted, or uses the supplied secret for manual setup. RevenueCat is webhook-only (no backfill): webhookSecret is the authorization header value configured on the RevenueCat webhook, or its HMAC signing secret. Superwall is webhook-only too and needs only webhookSecret: the whsec_ signing secret of the Superwall webhook pointed at the returned webhookUrl. Mixpanel and Amplitude need only webhookSecret, a strong random value you choose: Mixpanel sends it as the Basic auth password of its webhook destination (any username, the dashboard suggests sequenzy) and syncs cohorts as mixpanel-<cohort> tags; Amplitude sends it in an X-Webhook-Secret header and streams events (product.*), user properties, and cohorts (amplitude-<cohort> tags). Sequenzy cannot show the secret again, so give it to the user once. Attio is outbound-only and has no webhookUrl. SparkLoop needs only apiKey; Sequenzy registers the SparkLoop webhook itself with a secret in its URL (webhookSecret, the account's SparkLoop-Token, is optional), and an account with several publications needs providerAccountId (the 400 error lists them in publications). Reconnecting replaces stored credentials. OAuth and app-install providers (Stripe, Shopify, Supabase, GitHub, WooCommerce, Meta) are rejected with guidance: their flows need a human in the dashboard. SECURITY: only pass credentials the user explicitly provided for this purpose; suggest the CLI (`sequenzy integrations connect`) or dashboard when the user prefers keeping secrets out of the conversation.
get_integrationSolo lecturaInspect one connected integration in depth: what the provider syncs, every event it can emit and when, every matching tag rule, which sequences trigger on those events, recent integration-specific activity, and prioritized recommendations. For Webflow, `webflowCapture` reports the exact selected native forms, retained field names, double-opt-in policy, list IDs, sequence toggle, webhook status, and schema fetch time; capture is off when its forms array is empty. `observedByAccount` and `accountLastSeenAt` are account-wide because event definitions can also be updated by another integration or the public events API; use the retained activity log for connection-specific delivery diagnosis. For Attio, an `attio_lists_unmapped` recommendation means nothing is mapped so nothing syncs, and `attio_inbound_sync_error` means changes made in Attio are not reaching Sequenzy — use get_attio_mapping then update_attio_settings; do not reconnect for that. Read-only; credentials are never returned. Get the id from list_integrations.
list_integration_capabilitiesSolo lecturaDescribe what Sequenzy's integration providers do, whether or not they are connected. For each provider: category, how it connects, what it syncs, every event it emits with the real-world moment that triggers it, the subscriber attributes it writes, supported actions, availability, and caveats. Use this to answer 'what would connecting Stripe give me?', to compare providers before recommending one, or to check whether an event a sequence needs actually exists for that provider. Coverage differs sharply between providers - only Shopify has storefront browse tracking, and only Stripe, Chargebee, and RevenueCat cover the full trial lifecycle - so check here rather than assuming.
list_integration_suggestionsSolo lecturaSuggest integrations to connect next, based on what Sequenzy found on the company's website during onboarding: tools on the homepage or the sign-in page it links to (Clerk, Supabase, PostHog, Segment, Mixpanel, Amplitude, Webflow, Wix, WordPress, WooCommerce) or a Shopify storefront, minus anything already connected. Payment providers are never guessed from the site: when the website lists pricing and no payment or commerce integration is connected, `paymentProviders` lists every payment provider so the user can pick theirs. `scanComplete` is false while the website is still being read, so an empty list then means 'not scanned yet', not 'nothing found'. Use list_integration_capabilities to explain what a suggested provider gives and how it connects. Read-only.
list_integration_activitySolo lecturaRead recent integration webhook and sync activity, newest first. This is the log to check when an integration reports connected but contacts or events are not appearing: each row carries the action, status, event type, matched contact, and error. Payloads are sanitized at write time, so no credentials or secrets appear here. Activity is retained for 24 hours only.
set_integration_sync_enabledAcciónTurn a supported integration's future bulk imports and backfills on or off. THIS DOES NOT STOP AN INTEGRATION CREATING CONTACTS: disabling sync only pauses full imports and backfills, while the provider's live webhook keeps writing every new signup or customer as it happens. To change where those contacts land, use set_integration_list_targeting; to stop them arriving at all, use disconnect_integration for Lemon Squeezy or SparkLoop, or disconnect from the dashboard for other providers. Check get_integration.availableActions first; providers without enable_sync or disable_sync are dashboard-managed. Disabling keeps the connection, credentials, and live webhook delivery active, and requires any in-flight sync to finish first. It does not disconnect the integration. Idempotent: setting the current state succeeds with `changed: false`. Requires an API key with the `integrations:manage` scope, which agent-safe keys deliberately do not carry - without it, report the dashboard path (Settings -> Integrations) instead of retrying.
set_integration_list_targetingAcciónChoose which lists the contacts a connected integration creates join. This is the tool for 'stop this integration adding people to my marketing lists' - set_integration_sync_enabled only pauses bulk backfills and leaves the provider's live webhook writing. Scope: this changes list membership only, and it takes effect on future provider writes. It does not stop contacts being created, does not stop their attributes syncing, does not stop sync-rule tags, and does not stop `contact_added` sequences using the default `any_contact` scope, which fire precisely because the contact joined no list. Explicit `any_list` and specific-list sequences require a matching membership and do not enroll a list-less contact. Nobody is ever removed from a list and nothing is applied retroactively. Provider behavior for existing contacts differs: Wix or Webflow submissions, Shopify customer updates, and Supabase resubscriptions can add an existing contact to the new target lists; Stripe applies targeting only when its webhook creates a subscriber. Webflow is deliberately fail-closed: `null` is rejected, so pass `[]` or explicit list IDs; form selection, field allowlists, required double opt-in, and sequence enrollment remain dashboard-only under Settings -> Integrations -> Webflow. With no selected forms, Webflow capture is off and no contacts are created. If the goal is to stop `any_contact` enrollments too, pair this tool with pause_sequence_enrollments on the affected sequences. Check get_integration first - `ingestion` reports the current targeting with list names, and `availableActions` says whether this provider honors set_list_targeting. Supported for Supabase, Stripe, Shopify, Wix, Webflow, and SparkLoop. Dodo Payments, PostHog, Polar, Paddle, and other providers without that action reject this tool; their live contacts follow defaultSubscriberListIds on get_company/update_company instead (null = every list, [] = none), which is workspace-wide. PostHog history imports are an exception and create contacts without list memberships. Idempotent: asking for the current state succeeds with `changed: false`. Requires an API key with the `integrations:manage` scope, which agent-safe keys deliberately do not carry - without it, report the dashboard path (Settings -> Integrations) instead of retrying.
get_integration_pixelSolo lecturaCheck whether a Shopify store's storefront tracking pixel is installed and reporting to this account. Read live from Shopify on every call, because a merchant can remove the pixel without Sequenzy hearing about it. Check this before telling the user that browse abandonment, cart recovery, product views, collection views, or storefront search will work: without a live pixel those events never arrive, the sequences built on them never fire, and nothing anywhere reports an error. Returns `pixel.healthy`, the endpoint and signed configuration state, and `dependentEvents` - the event names that rely on this pixel. When `pixel.error` is null and `pixel.healthy` is false, those events are confirmed dark; an error means Shopify could not confirm their state. Fix a confirmed unhealthy state with activate_integration_pixel. Shopify only; other providers return a clear error.
activate_integration_pixelAcciónInstall the Shopify storefront tracking pixel on the connected store, or repoint an existing one at this account. This is what turns on product views, cart activity, and browse abandonment - the data every browse-abandonment and cart-recovery sequence depends on. Idempotent: an already-live pixel returns `changed: false`. Events start flowing on the next storefront visit and nothing is backfilled for the period the pixel was off, so run it before, not after, building the sequence. If the store granted an older permission set, this fails with a message naming the reconnect step rather than silently doing nothing. Shopify only.
sync_integrationAcciónQueue a manual re-sync for an integration: customers and revenue for a payment provider (Stripe, Polar, Paddle, Dodo, Lemon Squeezy, Creem, Chargebee, Whop), the user backfill for Supabase, or the event-history import for PostHog and Segment. Use this for Supabase before a campaign that depends on names or plan attributes - the live trigger only sends rows that change after it was installed, so existing users have no attributes until a backfill runs. The Supabase sync reads the project, schema, and table already configured for the integration and cannot be pointed at a different table; if none is configured yet it fails saying so. This is also the supported way to retry a failed PostHog or Segment history import: it restarts from the beginning using the stored credentials, and already-imported events dedupe, so a re-run cannot duplicate them. Returns immediately - poll get_integration to watch `syncStatus` and `lastSyncAt`, and list_integration_activity for per-row outcomes. Fails with a conflict if a sync is already queued or running; terminal BullMQ failures release history imports for retry. Providers whose backfill is not API-triggerable (Shopify products, Affonso affiliates) return a clear error naming the dashboard. Check `availableActions` on get_integration to see whether this is supported before calling.
get_attio_mappingSolo lecturaInspect a connected Attio integration's two-way sync: every Sequenzy list, every Attio people-list the stored token can write (id, api slug, name), the saved maps in both directions (listMap and tagMap push Sequenzy lists and tags to Attio lists; inboundListMap and inboundTagMap pull Attio lists into Sequenzy lists and tags), syncRemovals, whether company matching from work-email domains is on, and `inbound` (whether the Attio webhook is listening, and the error when it is not). Use it when Attio is connected but nothing is mapped, and call it before update_attio_settings so you have Attio list ids or slugs instead of asking the user to paste UUIDs. Reads Attio live on every call. Credentials are never returned. Attio only; other providers return a clear error. Check get_integration first — availableActions includes update_attio_settings when mapping can be saved without reconnecting.
update_amplitude_settingsAcciónChoose which events a connected Amplitude integration records, without the webhook secret (reconnecting would replace the secret and stop deliveries until the destination is updated). syncAllEvents false plus eventAllowlist records only those types. skipInternalEvents (on for new connections) drops session, [Amplitude] autocapture, and other $ events unless allowlisted; identify always passes. Applies to later deliveries; dropped events are not replayed. Amplitude only. Requires the `integrations:manage` scope; otherwise use the dashboard (Settings -> Integrations -> Amplitude).
update_attio_settingsAcciónSet the two-way sync of an already-connected Attio integration, using the stored access token, so changing mappings does not require connect_integration or the Attio access token again. Sequenzy → Attio: listMap (Sequenzy list id → Attio list) and tagMap (Sequenzy tag → Attio list) upsert the person and add them to the Attio list when they join the list or get the tag. Attio → Sequenzy: inboundListMap (Attio list → Sequenzy list id) and inboundTagMap (Attio list → Sequenzy tag) create or update the subscriber and add them to the list or tag when someone is added to the Attio list, which triggers list and tag automations. Saving an inbound map makes Sequenzy register the Attio webhook itself; that needs the webhook:read-write scope on the token, and the call fails with instructions when it is missing. Each map is a full replacement: pass the complete desired map, `{}` to clear it, or omit it to leave it unchanged. syncRemovals (default false) mirrors removals in every mapped direction; off keeps sync add-only. Optionally set syncCompanyFromDomain. Provide at least one field. Idempotent: asking for the current state succeeds with changed: false. Only changes after saving sync; existing members are not backfilled. Check get_attio_mapping first for Sequenzy list ids and live Attio list ids/slugs, and list_tags for tag names. Attio only. Requires an API key with the `integrations:manage` scope, which agent-safe keys deliberately do not carry; without it, the same settings are available in the dashboard at Settings -> Integrations -> Attio -> Sync settings.
get_event_sampleSolo lecturaRead the latest actual received event in the selected workspace, across all subscribers. Requires subscribers:read. Returns sample: null when none exists. Copy sample.properties into start_sequence_test_run customVariables, editing as needed. This read does not fire an event or change the selected test recipient. Unlike get_event_schema, this is recorded account data, with no age cutoff.
get_event_schemaSolo lecturaRead the published payload of a built-in event: a real example payload per provider, plus every property path with its type, the merge tag that resolves it, and a note wherever the example alone is ambiguous. Call this BEFORE writing {{event.*}} merge tags or event property filters - an unrecognized merge tag renders as an empty string rather than an error, so a guessed property name ships silently broken. It is also the only place that spells out the traps the sample cannot: money fields ending in `Cents` are minor units while `predictedLtv` is whole currency units, `price` is a preformatted display string that must not be parsed, `churnRisk` is a 0-95 percent and not a probability, and `orderId` is a string on Shopify but a number on WooCommerce. Omit eventName to list every documented event, or pass provider to get one provider's payload instead of all of them. Custom events are first-class and deliberately absent from this list: `documented: false` means no sample is published, never that the event name is invalid or will be rejected. Static reference data - it describes the event shape, not what this account has received. Use get_event_sample to read the latest actual received payload.
create_template_from_exampleAcciónRemix a public Sequenzy email gallery email into a new email template for this company. The new email keeps the example's exact layout and design; AI rewrites every piece of text for the company and its logo, brand color and links are swapped in. The example brand is never named and none of its copy, offers or claims are reused. Saved as an HTML template. Requires templates:write. Takes 5-20 seconds. Creates no template if generation fails (a company's first AI generation still saves its inferred email design direction); each successful call creates another template. Accepts a gallery email URL (https://sequenzy.com/email-examples/brands/{brand}/emails/{email}, or for an email in a sequence the sequence page URL ending in #email-{email}) or brand plus email slugs. For a whole sequence use create_sequence_from_example.
preview_template_from_exampleAcciónPreview how a public Sequenzy email gallery email would read for this company without saving any email (a company's first AI generation saves its inferred email design direction, like generate_email): the same rewrite as create_template_from_example (the example's layout, every word written for the company, its logo, brand color and links swapped in), returned as subject, previewText and HTML. Pass website to write it for another site's brand instead, such as a prospect or client; it is looked up like the gallery's public preview. Requires templates:write. Takes 5-20 seconds. Call create_template_from_example to keep it as a template. Accepts a gallery email URL or brand plus email slugs.
preview_sequence_from_exampleAcciónPreview the first 3 emails of a public Sequenzy email gallery sequence written for this company, without creating a sequence or email (a company's first AI generation saves its inferred email design direction, like generate_email). Each email keeps the example's layout with every word rewritten; returns subject, previewText, HTML and the original day for each, plus how many emails were left out. Pass website to write it for another site's brand instead. Requires sequences:write. Takes 10-30 seconds. Call create_sequence_from_example to create the draft sequence with every email. Accepts a gallery sequence URL or brand plus sequence slugs.
create_sequence_from_exampleAcciónClone a public Sequenzy email gallery sequence into a draft sequence for this company, with the example's trigger family (signup, abandoned cart, order placed, or inactive tag) and send timing, up to 12 emails (the first 12 unless stepNumbers picks which; each kept email keeps its original timing from the trigger). AI then writes every email in the company's brand in the background (usually 30-60 seconds); poll get_sequence until enrichmentStatus is complete. If this tool returns enrichmentStatus not_queued with a warning, the draft exists but writing did not start: open it in the dashboard and do not call this tool again. The sequence stays a draft and sends nothing until enabled. Requires sequences:write. Each call creates another sequence. Accepts a gallery sequence URL (https://sequenzy.com/email-examples/brands/{brand}/sequences/{sequence}) or brand plus sequence slugs.
list_email_referencesSolo lecturaList real emails from the Sequenzy email gallery to model a new email on, for the kind of email being written: campaign (promotions, newsletters, product updates), sequence (lifecycle emails plus whole sequences) or transactional. By default (scope similar) they come from the gallery brands most like the company: brands whose businesses (what each sells, and to whom) are closest in meaning to the company description, closest first (returned as brands; rankings are reused for up to 10 minutes). To rank them, the company description and any search text are sent to an embeddings service; nothing is saved. Each item has a gallery url that create_template_from_example (or create_sequence_from_example for a sequence) accepts, and emails carry an analysis (main button, length, layout, voice, structure and why it works; null until analyzed) to choose a model by. With scope all it browses the whole gallery instead; search narrows either. profile is missing (and the list empty) when the company has no description or company context yet; use scope all then. Requires templates:read.
list_gallery_brand_requestsSolo lecturaList this company's requests to add brands (usually competitors) to the Sequenzy email gallery, newest first. Each has a status: requested (not added yet), collecting (the brand is added and its emails are being collected), available (its emails are in the gallery; brand.url links to them and list_email_references can return them) or declined (with declineReason). Requires templates:read.
request_gallery_brandAcciónAsk for a brand, usually a competitor, to be added to the Sequenzy email gallery so its real emails can be used as references. Pass its website or domain. Asking again for the same domain keeps the one request (a new note replaces the old) and returns created false. If the brand is already in the gallery, the request comes back with status available. A company can keep up to 50 requests. It is not your own website. Requires templates:write.
remove_gallery_brand_requestAcciónWithdraw one of this company's brand requests by id (from list_gallery_brand_requests). The brand stays in the gallery if it was already added. Requires templates:write.
list_watched_brandsSolo lecturaList the brands (usually competitors) this company watches, newest watch first. Each has a status: available (its emails are in the gallery; emailCount and latestEmailAt say how many and how recent), requested (not in the gallery yet), collecting (added, emails on the way) or declined (with declineReason). Use list_watched_brand_emails to read what they sent. Requires templates:read.
watch_brandAcciónWatch a brand, usually a competitor, by its website or domain. A brand already in the Sequenzy email gallery is watched at once (status available) and its emails can be read with list_watched_brand_emails; any other brand is requested and its emails show up once it is added and they are collected. New emails arrive a few hours after they are sent. Watching again keeps the one watch (a new note replaces the old) and returns created false. A company can keep up to 50 watches and brand requests together. It is not your own website. Requires templates:write.
unwatch_brandAcciónStop watching a brand by its watch id (from list_watched_brands). This also withdraws the brand request with the same id. The brand stays in the gallery. Requires templates:write.
list_watched_brand_emailsSolo lecturaRead emails the watched brands sent, newest first, a page at a time: subject, type, subtype, sentAt, collectedAt, brand, url and a breakdown (analysis, null until analyzed). Pass domain for one brand and since for emails sent at or after a time (for example the start of this week). Emails arrive a few hours after they are sent, so when polling set since about a day before the last check and skip ids already seen. Pass nextCursor back as cursor for older emails. Each url also works as the url of create_template_from_example to remix it. Requires templates:read.
start_subscriber_operationAcciónAdd tags to a whole filtered audience in the background. Returns a durable operation ID; poll get_subscriber_operation or list_subscriber_operations. Reuse requestKey after an uncertain response; changing inputs with the same key is rejected. Tagging needs subscribers:tag, new definitions need tags:write, and triggerAutomations needs automations:trigger. Existing synchronous bulk tag tools remain suitable for up to 500 explicit contacts.
list_subscriber_operationsSolo lecturaList the company's 20 most recent background tagging operations, including progress and failures.
get_subscriber_operationSolo lecturaRead background operation progress and state. A completed tagging operation may have per-contact failures; inspect failed and failures.
cancel_subscriber_operationAcciónStop future work for a queued/running tagging operation. Already applied tags are retained. Tagging cancellation requires subscriber tagging access.
add_subscriberAcciónAdd one subscriber. Status is only applied when creating a new contact; use update_subscriber to change an existing contact's status. The subscribers:write scope covers listIds in this call. Matching sequences enroll by default (REST enrollInSequences), including tag_added when tags are set in this call, without requiring a list; pass enrollInSequences false to skip, and automations:trigger is required for enrollment. Without automations:trigger the call succeeds but starts no sequence (sequenceTriggersRan false, reason missing_scope). For multiple contacts or CRM records, use create_subscriber_import instead of looping this tool. For an email-only batch going into one list, add_subscribers_to_list is also available.
create_subscriber_importAcciónQueue one CRM-grade subscriber import with up to 5,000 full contact records. Supports names, external IDs, phones, statuses, tags, lists, and custom attributes. Prefer this over repeated add_subscriber calls. Returns an import ID for get_subscriber_import. Use optInMode='confirmed' only when the contacts already gave verified email consent. Rows are imported only after a list check, usually under a minute. When the list has too many addresses that don't exist or look risky (temporary inboxes, or mostly shared addresses like info@), nothing is imported and the finished import has listQuality (status 'needs_cleaning', with issues): relay listQuality.message and suggest cleaning the list (listQuality.cleaningUrl) and importing the cleaned list.
get_subscriber_importSolo lecturaGet progress, counts, skipped reasons, and failure summaries for a queued subscriber import. Every excluded row is explained: skippedReasons sums to skippedCount and failedReasons sums to failedCount, so a finished import can be reconciled without inspecting individual contacts. If the reason counts fall short of the totals, report it with the import ID rather than guessing what the missing rows were. listQuality is null unless the list needs cleaning; then nothing was imported (rows skipped as list_needs_cleaning) and you should relay its message and cleaningUrl. An import stays running until its list check finishes.
update_subscriberAcciónUpdate an existing subscriber's profile, phone, status, attributes, or tags. Names and phones are native profile fields - set them here rather than as custom attributes. Setting status to unsubscribed suppresses the contact from all marketing, deactivates list memberships, and cancels active sequence enrollments while preserving the record and suppression history. Without automations:trigger, added tags start no sequence (sequenceTriggersRan false).
remove_subscriberAcciónUnsubscribe a subscriber while preserving the record and suppression history: the contact is suppressed from all marketing, list memberships are deactivated and active sequence enrollments are cancelled. With hardDelete true the contact is permanently deleted and its history erased, which cannot be undone. Identify the contact by email or externalId.
get_subscriberSolo lecturaGet the full subscriber profile, including tags, notes, list memberships, sequence enrollments, email stats, and recent activity. For commerce customers, customAttributes include nightly predictive analytics: predictedLtv (next-12-month spend), churnRisk (percent), expectedNextOrderAt, avgDaysBetweenOrders, and predictionConfidence.
list_subscriber_notesSolo lecturaList internal notes for a subscriber. Provide email or externalId to identify the subscriber.
list_subscriber_attributesSolo lecturaList the custom attribute names in use across the account, each with its value type (string, number, or boolean), whether it holds a list (isArray), an example value, and how many of the sampled recent contacts carry it. Call this before add_subscriber or update_subscriber to reuse existing names and match their types, e.g. send a zip code as the string "02134" rather than a number. Names only older contacts carry are included with sampledContacts 0. mixedTypes is true when contacts hold different types for a key; valueType is then string. Always send values with leading zeros as strings, whatever valueType says.
add_subscriber_noteAcciónAdd an internal note to a subscriber. Provide email or externalId to identify the subscriber.
delete_subscriber_noteAcciónDelete one internal subscriber note by note ID.
trigger_subscriber_eventAcciónEmit a custom event for one subscriber, exactly as an integration or the public API would. This is the supported way to exercise event triggers, matching-field idempotency, branch conditions, and stop conditions end to end without waiting for real traffic. The event is recorded, sync rules apply, and matching event_received sequences enroll. Without automations:trigger the call succeeds but starts no sequence (sequenceTriggersRan false, reason missing_scope). Creates the subscriber if they do not exist. Use trigger_subscriber_events for several events on the same contact.
trigger_subscriber_eventsAcciónEmit several custom events for one subscriber in order, exactly as an integration would: each event is recorded, sync rules apply, and matching event_received sequences enroll. Creates the subscriber if they do not exist. Events are processed independently and sequentially, so a partial failure can still leave earlier events recorded; retry with the same eventIds. Returns the subscriber and the recorded events. Without automations:trigger the call succeeds but starts no sequence (sequenceTriggersRan false, reason missing_scope).
import_subscriber_eventsAcciónRecord a bounded batch of events for many subscribers in one call, built for data pipeline and warehouse syncs. Each event carries its own email or externalId and a required source-owned eventId; externalId-only rows must identify an existing contact because email is required for creation. Contacts whose events are all more than an hour old are imported silently as history (including no double-opt-in email); any recent event makes that contact's whole group live. Retries keep one receipt and re-attempt any downstream recovery that has not finished. Accepts at most 25 events. Without automations:trigger the call succeeds but starts no sequence for live rows (sequenceTriggersRan false, reason missing_scope).
bulk_add_subscriber_tagsAcciónMany contacts, one or more tags: add tags to up to 500 existing subscribers in one call, identified by emails, externalIds, or subscriberIds. Never creates contacts. Built for reconciling historical or derived tags: subscribers that do not exist are reported in notFound instead of being created, and tag automations do NOT run unless triggerAutomations is true. Returns the normalized tags and counts of requested, matched, updated, unchanged and failed subscribers, with notFound identifiers and up to 50 failures. Needs the subscribers:tag scope; a tag name that does not exist yet also needs tags:write. The agent_safe key preset includes both. Use add_subscriber or update_subscriber for single-contact changes.
bulk_remove_subscriber_tagsAcciónMany contacts, one or more tags: remove tags from up to 500 existing subscribers in one call, identified by emails, externalIds, or subscriberIds. Never creates contacts: subscribers that do not exist are reported in notFound. Needs the subscribers:tag scope; the broader subscribers:write scope also satisfies it. Use this to roll back or reconcile a derived-tag backfill. Returns the normalized tags and counts of requested, matched, updated, unchanged and failed subscribers, with notFound identifiers and up to 50 failures.
search_subscribersSolo lecturaSearch subscribers by free-text query, tags, list, segment, custom attribute, or opt-out date. Each subscriber carries unsubscribedAt, the date the contact opted out - use it rather than updatedAt, which any later tag or attribute write overwrites. If you omit limit, the tool fetches all pages and returns every match in one call - that is the fastest way to enumerate an audience. To walk a large audience in chunks instead, pass limit and then follow pagination.nextCursor (or pagination.nextOffset) until pagination.hasMore is false. Default searches are newest first; searches with attribute use stable subscriber-ID ascending order. Returns the matching subscriber records, the count returned, and pagination with total, hasMore, nextCursor and nextOffset.
list_accountsSolo lecturaList your customers' organizations (B2B accounts such as companies, workspaces or teams) in the workspace. The dashboard shows them under Subscribers > Organizations; these tools keep the account names. Not your Sequenzy account, workspace or login: use get_account for that. Search matches name, external ID and domain. Organizations group contacts; their attributes reach every member as `organization.<name>` for segment attribute filters and sequence conditions, and as `{{organization.<name>}}` merge tags (`account.*` is an older alias).
get_account_by_external_idSolo lecturaGet one customer organization (Subscribers > Organizations in the dashboard) by its external ID, including up to 100 member contacts with their roles (owner, admin, member). Not your Sequenzy account: use get_account for that.
upsert_accountAcciónCreate or update a customer organization (Subscribers > Organizations in the dashboard) by external ID. Attributes merge into the stored attributes (a null value deletes a key; set replaceAttributes to replace them all). Optional members are added in the same call. Use this to model companies, workspaces or teams so segments can filter on `organization.plan` and emails can use `{{organization.name}}` (`account.*` still works). The first organization turns Organizations on for the workspace; it fails with a 409 error when contacts already use a custom attribute named `account` or `account.*`, which must be renamed first.
delete_accountAcciónDelete one customer organization (a B2B account your contacts belong to, shown under Subscribers > Organizations) and its memberships. Contacts are kept; their `organization.*` attributes are cleared. It never deletes the user's Sequenzy account, workspace (company), login or API keys. If the user asks to delete their own Sequenzy account or workspace, do not call this tool: that is only possible in the Sequenzy dashboard. Permanent; confirm the organization's externalId with the user first.
list_account_membersSolo lecturaList the contacts that belong to a customer organization with their roles. Owners first, then admins, then members.
add_account_memberAcciónAdd a contact to a customer organization (creating the contact when the email is new). Re-adding an existing member with a role updates the role; without a role keeps the current one.
remove_account_memberAcciónRemove a contact from a customer organization. The contact is kept; its `organization.*` attributes are cleared or replaced by another organization it belongs to.
trigger_account_eventAcciónRecord an event on a customer organization (for example trial_ending, plan_changed, seat_limit_reached) and deliver it as a contact event to the chosen recipients: owners (default; falls back to admins when there is no owner), admins (includes owners), all members, or none (timeline only). Each delivery runs sync rules and sequence triggers with `event.organization.*` properties (also available as `event.account.*`). To start a sequence once per organization rather than once per recipient, create it with enrollmentMode matching_field and enrollmentFieldPath organization.externalId. Pass eventId to make retries idempotent.
list_account_eventsSolo lecturaList recent events on a customer organization's timeline, newest first.
list_account_suggestionsSolo lecturaSuggest accounts from work email domains that several contacts share (for example 7 contacts at acme.com). Personal and disposable providers (gmail.com, outlook.com, yopmail.com, ...), the workspace's own sending domains, domains where any contact already belongs to an account, and domains an account already uses are skipped. Each suggestion includes up to 10 sampleEmails for review. Read-only; show the list to the user and accept only the domains they confirm with accept_account_suggestions.
accept_account_suggestionsAcciónCreate accounts from suggested email domains. Each domain becomes an account (externalId and domain set to the domain, name derived from it) with the domain's contacts that belong to no account as members. When one account already uses the domain, contacts are added to it; with several, the domain is skipped (multiple_accounts). A domain whose contacts already partly belong to another account is skipped (already_in_account); domains without eligible contacts are skipped (no_contacts). Up to 25 domains per call; personal, disposable and the workspace's own sending domains are rejected. Keeps existing roles, runs no sync rules or segment-entered sequences, and a retry does not duplicate accounts or members; truncated means contacts were left out because an account holds at most 5,000 members. Fails with a 409 error while contacts use a custom attribute named `account` or `account.*`.
detect_account_organization_idsSolo lecturaFind the organization ID the app already sends, so accounts can be created from existing data without code. Looks at event properties from the last 180 days and contact attributes named like workspaceId, organization_id, team_id or account_id. Each candidate has source (event or attribute), propertyKey, nameKey (a sibling name property or null), eventCount (events for event, contacts for attribute), accountCount (distinct organizations) and eventNames. Read-only; next, preview a candidate with preview_accounts_from_organization_id and show it to the user.
preview_accounts_from_organization_idSolo lecturaPreview the biggest accounts create_accounts_from_organization_id would make, before anything is written. Each organization has externalId (its ID), name (from nameKey, otherwise from the work email domain most of its contacts share, null when neither exists), domain, contactCount and up to 3 sampleEmails. Read-only; always show this preview to the user and get confirmation before creating accounts.
create_accounts_from_organization_idAcciónStart a background job that creates one account per distinct propertyKey value and adds each contact seen with it as a member. Accounts are named from nameKey, otherwise after the work email domain most of their contacts share; existing names, domains and roles are kept. No emails are sent, no sync rules run and no segment-entered sequences start. Returns jobId, alreadyRunning (a run for the same key was already queued or running, so none was started) and job; check progress with get_account_organization_id_job. Reruns update the same accounts instead of duplicating them. Only call after the user confirmed the preview. Fails with a 409 error while contacts use a custom attribute named `account` or `account.*`.
get_account_organization_id_jobSolo lecturaCheck a create_accounts_from_organization_id run. state is queued, running, completed or failed. A completed job has result with accountsCreated, accountsUpdated, membershipsCreated, membershipsUpdated, skipped and truncated (history was larger than one run covers; start it again to continue). A failed job has error. Finished jobs expire after a while (404); the accounts stay.
list_recipient_suppressionsSolo lecturaList the recipients this company cannot reach: protected global invalid-recipient suppressions for addresses associated with the company, protected workspace hard bounces without conclusive invalid-inbox evidence, removable workspace soft-bounce escalations, and protected spam complaints. Each row includes a stable suppressionType, and the response echoes the applied sortBy/sortOrder. Sort by `status` to surface the suppressions you can actually remove. Start here when a customer reports missing email; use get_recipient_suppression for one exact address.
get_recipient_suppressionSolo lecturaCheck whether one recipient is suppressed by Sequenzy's bounce/complaint safeguards. Suppression comes solely from Sequenzy's own bounce records; the email provider's suppression list is not consulted. This exact-address lookup never exposes other recipients.
remove_recipient_suppressionAcciónRemove a workspace-scoped soft-bounce escalation for an exact recipient associated with the company, then reactivate bounced company subscribers. Global invalid-recipient, workspace hard-bounce, complaint, and unsubscribe suppressions are protected.
list_productsSolo lecturaList synced products (Stripe, Shopify, WooCommerce) including any attached digital delivery file. Useful before attaching a file or building a purchase sequence for a specific product. For Stripe products, variants are the product's active prices (variantId = Stripe price ID), so an exact price can be targeted even when it is not the default. Returns one page: check pagination.total and pagination.hasMore, and page with offset to read the whole catalog.
upsert_productsAcciónCreate or update products in the catalog (Commerce API, keyed by your productId, stored under the api provider). Use this to add products that are not synced from Stripe/Shopify/WooCommerce, then attach a deliverable file with attach_product_file and build a purchase sequence for them.
delete_productAcciónDelete a product previously pushed via upsert_products (Commerce API products only), identified by your productId.
attach_product_fileAcciónAttach a distributable file to a product, either by public URL or by uploading a local file (filePath, local MCP server only). After a purchase of the product, sequence emails can link to it with {{event.download.url}} and {{event.download.name}}.
remove_product_fileAcciónRemove the attached distributable file from a product.
sync_productsAcciónQueue a sync of the Stripe product catalog into the products list. Requires an active Stripe integration with bulk sync enabled. Pass integrationId when more than one Stripe account is connected; otherwise the most recently connected enabled account is used.
get_push_settingsSolo lecturaGet push notification readiness: whether web push is enabled (with its VAPID public key), whether APNs (iOS) and Firebase (Android) credentials are connected, the last credential error for each, and active device counts per platform. Check this before creating push campaigns or push sequence steps.
update_web_push_settingsAcciónEnable or disable web push, or set the default notification icon. The first enable generates the workspace's VAPID keys; disabling keeps them so existing browser subscriptions stay valid. The site must serve /sequenzy-sw.js (importScripts of the hosted worker) and call sequenzy.subscribeToPush() from a click.
set_apns_credentialsAcciónConnect iOS push by saving an APNs auth key (.p8). The key is encrypted at rest and never returned. Requires the integrations:manage scope and a role that can edit workspace settings (marketer keys cannot).
remove_apns_credentialsAcciónRemove APNs credentials. iOS devices stop receiving push until a key is saved again; registered devices are kept.
set_fcm_credentialsAcciónConnect Android push by saving a Firebase service account key (the JSON file from Firebase project settings > Service accounts). Encrypted at rest and never returned.
remove_fcm_credentialsAcciónRemove Firebase credentials. Android devices stop receiving push until a key is uploaded again; registered devices are kept.
list_push_devicesSolo lecturaList registered push devices (browsers and app installs) in a stable order; pass nextCursor to page. Filter by contact (email or subscriberId), platform or status. Tokens are never returned in full.
register_push_deviceAcciónRegister (or refresh) a push device for a contact, the way an app backend would after the app receives its token. Re-registering a token reactivates it and moves it to the given contact. Unknown emails become contacts quietly (no lists, no contact_added sequences).
remove_push_deviceAcciónUnsubscribe a push device and unlink it from its contact so it stops receiving push. A browser or app that still allows notifications can register again (browsers come back unlinked).
send_test_pushAcciónSend a real test push to one device (deviceId) or to every active device of a contact (email or subscriberId). Provide title and/or body. Limited to 200 per company per rolling 24 hours; excluded from stats.
list_push_campaignsSolo lecturaList push campaigns, newest first.
get_push_campaignSolo lecturaGet a push campaign's content, audience, status and (after sending) delivery stats.
create_push_campaignAcciónCreate a push campaign draft. Content and audience can be set now or later with update_push_campaign; send with send_push_campaign.
update_push_campaignAcciónUpdate a draft or scheduled push campaign. Only the fields you pass change. A scheduled campaign keeps its schedule; timing fields can only change on drafts (reschedule with send_push_campaign instead).
estimate_push_campaign_recipientsSolo lecturaCount how many contacts in an audience have an active push device (eligibleCount) versus the whole audience (audienceCount).
send_push_campaignAcciónSend a push campaign now, or schedule it with scheduledAt (future ISO 8601). Optionally deliver in each contact's local time (sendInRecipientTimezone + scheduledTimezone) or at each contact's optimal hour (sendTimeOptimization); omitted timing fields keep the draft's saved values. Delivers real notifications to every contact in the audience with an active device on a set-up platform. Confirm with the user before calling.
unschedule_push_campaignAcciónMove a scheduled push campaign back to draft.
cancel_push_campaignAcciónStop a scheduled or sending push campaign. Unsent notifications are skipped; delivered ones cannot be recalled.
duplicate_push_campaignAcciónCopy a push campaign (content and audience) into a new draft.
get_push_campaign_statsSolo lecturaGet push campaign delivery stats: sent, displayed (device-reported), clicked, failed and skipped contacts, devices targeted/accepted, and skip/failure reasons.
translate_push_campaignAcciónTranslate a draft or scheduled push campaign's title and message with AI into the workspace's languages and save the versions. Contacts then get the version for their language attribute or device language. Versions a person wrote are kept unless overwrite is true. Returns the campaign plus translated, skipped and failed languages.
translate_push_contentSolo lecturaTranslate a push title and message with AI into the workspace's languages without saving anything. Use the returned localizations for a sequence push step (update_sequence_node changes.localizations, or localizations on a push step in create_sequence/insert_sequence_step) or a push campaign.
get_push_import_preflightSolo lecturaCheck push settings before importing devices: warns when APNs or Firebase is not set up, or APNs is on sandbox. Imported tokens only work with the same bundle ID, Apple team and Firebase project the app used with the previous provider. Show these warnings to the user before starting an import.
list_push_importsSolo lecturaList recent push imports with status, progress and counts (imported, matched, unsubscribed, skipped, invalid, contacts created).
get_push_importSolo lecturaGet one push import: status (draft, queued, exporting, importing, completed, failed, canceled), progress, counts, skip/invalid reasons, warnings and whether it can be resumed. Poll this while it runs.
create_push_importAcciónCreate a draft push import and return its warnings; nothing is read until start_push_import. Requirements by source: onesignal needs apiKey (App API key) and onesignalAppId; braze needs apiKey (REST key with users.export.segment), brazeRestEndpoint and brazeSegmentId; customer_io needs apiKey (App API key), customerIoSegmentId and optionally customerIoRegion; csv takes the file as csv (token and platform columns, plus external_id or email; at most 5 MB, send more with add_push_import_rows and set csvParts to the total part count so a partial upload can never start). Opted-in devices are linked to contacts by external ID or email (unmatched devices get a new contact, quietly: no lists, no sequences; push-only with no email or phone when the device has only an external ID); opted-out devices are kept as unsubscribed.
add_push_import_rowsAcciónAdd another part of a CSV file to a draft csv import (at most 5 MB per part; every part starts with the header row). Parts are imported in the order they are added.
start_push_importAcciónStart a draft push import, or resume a failed or stalled one from where it stopped. Runs in the background; poll get_push_import. One import runs per workspace at a time.
cancel_push_importAcciónStop a push import, or discard a draft or failed one. Devices already imported stay; staged rows and the stored source key are deleted. Cannot be undone.
list_live_activitiesSolo lecturaList iOS Live Activity and Android Live Update tokens (push-to-start tokens per device and activity type, update tokens per running activity) with a per-type summary of devices that can start activities and running activities. Each token has platform ios or android. Tokens are masked.
start_live_activityAcciónStart a Live Activity remotely on the devices that registered a push-to-start token for activityType: iOS 17.2+ through APNs, and Android Live Updates through FCM on apps that observe the same type name with the Sequenzy Android SDK (promoted on Android 16, an ongoing notification before). Target one contact (email or subscriberId), or every contact in a segment, list or tag (exactly one of segmentId, listId, tag; up to 100,000 devices, needs campaigns:send), then update and end it with broadcast: true. Requires an alert. Returns tokenCount 0 without sending when no device can start it; contactCount and platforms count contacts and devices per platform. Repeating the call starts another copy.
update_live_activityAcciónUpdate every running copy of a contact's Live Activity (one per device, iOS and Android Live Updates) with new content state. Requires email or subscriberId, or broadcast: true to update every signed-in device running the activityId (for activities started for a segment, list or tag). Add an alert to light the screen. Returns tokenCount 0 when no running copy has reported an update token.
end_live_activityAcciónEnd a contact's Live Activity on every device running it (iOS and Android Live Updates), with its final content state and optionally a dismissal date. Requires email or subscriberId, or broadcast: true to end it on every signed-in device running the activityId. Without dismissalDate iOS keeps it on the Lock Screen and Android in the shade for up to four hours.
list_in_app_messagesSolo lecturaList a contact's in-app messages and content cards (created by type "in_app" sequence steps). status "open" (default) is what their app should show now; "all" is history. Identify the contact by email, externalId or subscriberId.
record_in_app_eventAcciónRecord that the app displayed, clicked, dismissed or read an in-app item. Repeats change nothing except clickCount. Clicks, dismissals and reads also count as a display. Messages leave the open feed once displayed, clicked or dismissed; cards once dismissed.
delete_in_app_messageAcciónDelete an in-app message or card so the app stops showing it.
get_in_app_step_statsSolo lecturaCreated, displayed, clicked and dismissed counts for one in-app sequence step, plus how many contacts allowed (pushPermissionGranted) or declined (pushPermissionDenied) notifications from a push primer.
list_in_app_templatesSolo lecturaStarting points for in-app sequence steps (trial ending, welcome, new feature, milestone, a push primer that asks for notification permission, and more). Each has a complete step config: pass it as a type "in_app" step to create_sequence or insert_sequence_step after adding the urls its link buttons open.
validate_in_app_htmlSolo lecturaCheck a custom HTML in-app message document before using it as a format "html" in-app step. Returns its size against the 100 KB limit, Braze Liquid conversions (e.g. {{${email_address}}} becomes {{email}}, {{${braze_id}}} becomes {{contactId}}), script hosts a website CSP must allow, whether it closes itself, warnings and blocking errors. Nothing is saved.
write_in_app_copySolo lecturaWrite in-app message copy with AI from a short brief: eyebrow, title, message and button labels sized for the layout. Returns text only; nothing is saved. Button links are never invented, so add urls before using the copy in a step.
list_in_app_campaignsSolo lecturaList in-app campaigns, newest first. Each has a stored status and a state: draft, scheduled, live, ended (past endsAt) or stopped.
get_in_app_campaignSolo lecturaGet an in-app campaign's message, audience, priority, schedule, end time and state, plus reach and engagement stats once it has launched.
create_in_app_campaignAcciónCreate an in-app campaign draft: one in-app message or content card for an audience, shown inside the customer's app. Set the message and audience now or later with update_in_app_campaign; launch with send_in_app_campaign.
update_in_app_campaignAcciónUpdate a draft or scheduled in-app campaign. Only the fields you pass change; message fields merge over the stored message. Live campaigns can't be edited: stop and duplicate them instead.
estimate_in_app_campaign_recipientsSolo lecturaCount contacts who match an audience now (audienceCount) and how many are known to have your app connected (appUserCount, a lower bound). Everyone in the audience gets the campaign the next time their app fetches the feed, including contacts who join while it is live. If appUserCount is 0, say Sequenzy hasn't seen their app connect yet; if in-app isn't set up, point to Settings > In-app.
send_in_app_campaignAcciónLaunch an in-app campaign now, or schedule it with scheduledAt (future ISO 8601). While live, every contact in the audience sees the message the next time their app fetches its feed. Confirm with the user before calling.
unschedule_in_app_campaignAcciónMove a scheduled in-app campaign back to draft.
stop_in_app_campaignAcciónStop a scheduled or live in-app campaign. Nobody new gets it and it leaves every feed that still has it, including seen cards; stats are kept. Stopping is final (duplicate to run it again); stopping again succeeds.
duplicate_in_app_campaignAcciónCopy an in-app campaign (message, priority, audience and a future end time) into a new draft.
get_in_app_campaign_statsSolo lecturaGet in-app campaign results: delivered (contacts whose app received it), displayed, clicked, totalClicks, dismissed, push permission results for primer buttons, and first clicks per custom HTML button, plus the audience size at launch.
list_in_app_placementsSolo lecturaList the workspace's in-app placements: named slots in the customer's app or website that show in-app content inline (key, name, description, size hints) with all-time counts (created, live, displayed, clicked, dismissed). Fill one from a sequence step with type "in_app", kind "placement" and placement set to its key.
get_in_app_placementSolo lecturaGet one in-app placement by key, with its all-time counts: items created, live (open now), and contacts who displayed, clicked or dismissed their item.
create_in_app_placementAcciónCreate an in-app placement: a slot in the customer's app that shows in-app content inline, such as a home screen hero. The key is what app code passes to the SDK and never changes. Up to 50 per workspace.
update_in_app_placementAcciónRename an in-app placement or change its description or size hints. The key cannot change. Provide at least one field.
delete_in_app_placementAcciónDelete an in-app placement. Its slots collapse in the app on the next refresh, and in-app steps that target it skip contacts until a placement with the same key exists again. Items already created stay in contact history.
get_in_app_placement_contentSolo lecturaWhat each placement shows one contact now: for each key, the contact's highest-priority open placement item (ties: newest), or null when the slot collapses. Identify the contact by email, externalId or subscriberId. Reading marks nothing displayed.
list_tagsSolo lecturaList every tag defined in the company, with each tag's ID, normalized name and color. Use it to reuse existing tag names before tagging contacts, or to find a tagId.
create_tagAcciónCreate a new tag definition. The name is normalized to lowercase with hyphens (e.g. 'VIP Customer' becomes 'vip-customer'). Color defaults to gray. Returns the created tag with its ID, normalized name and color. Creating a tag does not apply it to any contact.
update_tagAcciónUpdate a tag's color. System tags cannot be updated.
delete_tagAcciónPermanently delete a tag and remove it from all subscribers. This cannot be undone. System tags and tags used by sequences cannot be deleted.
list_listsSolo lecturaList all subscriber lists. Each list includes subscriberCount (current members of any status) and activeSubscriberCount (current members with status=active). Members who unsubscribed from the list are not counted. Use these fields for list audience size instead of paging search_subscribers.
create_listAcciónCreate a new subscriber list. Lists are public by default: public lists appear to subscribers in the email preference center, where they can join or leave them. Set isPrivate to keep the list internal.
update_listAcciónUpdate a subscriber list's name, description, or privacy. Only the provided fields are changed. A public list's name is shown to subscribers in the email preference center, and making a list public lets subscribers join it there.
delete_listAcciónPermanently delete a subscriber list and remove all of its memberships. Subscribers themselves are kept. This cannot be undone.
add_subscribers_to_listAcciónBulk add subscribers to a subscriber list from an email array. Existing subscribers are added to the list without a per-subscriber update call; addresses not yet in the account are created as new subscribers, using optInMode for their consent. Returns the list plus added and skipped counts. Get listId from list_lists or create_list.
remove_subscribers_from_listAcciónRemove subscribers from a list by email address. Maximum 500 emails per call. Subscribers stay in the account; only the list membership is removed. Returns the removed count plus a notFound array of emails that did not match a subscriber.
list_segmentsSolo lecturaList every saved segment in the company with its ID and name. Use the IDs with get_segment_count, search_subscribers segmentId, or campaign targeting.
create_segmentAcciónCreate a new segment from explicit filter rules. Use `filters` plus `filterJoinOperator` for flat legacy rules, or `root` for nested AND/OR groups such as `{ "kind": "group", "joinOperator": "and", "children": [{ "kind": "filter", "field": "attribute", "operator": "gte", "value": "mrr:50" }, { "kind": "group", "joinOperator": "or", "children": [{ "kind": "filter", "field": "tag", "operator": "contains", "value": "vip" }, { "kind": "filter", "field": "event", "operator": "is_not", "value": "saas.purchase:30d" }] }] }`. AND of two or more positive attribute filters on the same array path (e.g. `history_events[].eventvenue_id` and `history_events[].showing_date`) requires one array element to satisfy every condition — a 2010 visit at that venue plus an unrelated recent booking elsewhere does not match. Negative filters (`is_not`, `not_contains`, `is_empty`) are exempt: they assert no element matches and combine independently. Supports `event` and `segment` fields, Stripe product purchase/current/trial/date filters, and campaign- or delivery-policy-scoped engagement filters such as `emailOpened is marketing:all`.
update_segmentAcciónUpdate a segment's name and/or filter rules. Use the same `filters` plus `filterJoinOperator` or nested `root` shapes as create_segment; providing `filters` or `root` replaces the segment's existing rules.
delete_segmentAcciónPermanently delete a segment. This cannot be undone. Subscribers are not affected. Deleting is never blocked when the segment is in use: the result lists `affectedSequences` (with how each uses it), `affectedCampaigns`, `affectedSegments` and `affectedAudienceSyncs`, which from then on treat every contact as not in the segment. Requires the segments:delete scope; segments:write is not enough.
get_segment_countSolo lecturaCount the subscribers who currently match a saved segment's rules. Returns count and the segmentId that was counted; it does not list the subscribers (use search_subscribers with segmentId for that).
list_audience_syncsSolo lecturaList Meta custom audience syncs (segment -> ad audience mappings) with their schedule and last sync status. Requires the Meta Ads integration to be connected in the dashboard.
list_ad_accountsSolo lecturaList the Meta ad accounts available through the connected Meta Ads integration. Use the returned account id as `adAccountId` when creating an audience sync.
create_audience_syncAcciónPush a segment to a Meta custom audience and keep it synced on a schedule. Provide `segmentId` for an existing segment OR `predefinedSegmentId` for a ready-made template (for example "zero-ltv", "no-purchase-1y", "recent-buyers", "high-spenders-ecom", "non-buyers", "engaged") - the template segment is created automatically on first use. The first upload runs immediately. Note: audiences are add-only; subscribers who later leave the segment stay in the Meta audience.
update_audience_syncAcciónUpdate an audience sync's frequency or pause/resume it (`isActive`).
delete_audience_syncAcciónRemove an audience sync mapping. The Meta audience itself is kept (running ads are not disrupted) - only future syncs stop.
sync_audience_nowAcciónTrigger an immediate upload of the segment's subscribers to its Meta custom audience, outside the regular schedule.
list_templatesSolo lecturaList the company's email bodies newest first, including per-locale localization sync status and each body's `emailPreset` Style > Format (branded or minimal; null for standalone raw HTML). Results include standalone templates plus the bodies behind campaigns and transactional emails. Returns 50 per call by default (100 max); page with offset while pagination.hasMore is true, and read pagination.total for the full count. You can also pass a campaign's emailId to get_template to fetch one body directly. Bodies are kept when their campaign or transactional email is deleted.
get_templateSolo lecturaGet a template's details, content, and all localized variants with sync status. `emailPreset` reports the per-email Style > Format (branded or minimal; null when the whole email is standalone raw HTML) the same way get_sequence does per step.
create_templateAcciónCreate a new email template. `prompt` generates net-new email content from natural language. `blocks` is for finished caller-supplied Sequenzy content and `html` for supplied or explicitly requested preserved HTML. Creating a standalone copy of a saved design and AI rewriting within its layout are currently dashboard-only workflows; this tool has no source-design operation, and `prompt` does not preserve an existing layout. For a campaign copy, use create_campaign with templateId.
update_templateAcciónUpdate an existing template in place: the supplied fields replace the saved ones (labels replace the whole label set). Campaigns and sequence steps that started from this design keep their own copies; automatically translated variants are refreshed from the new original unless marked keepEdits (see set_template_localization). At least one of `name`, `subject`, `previewText`, `html`, `blocks`, `labels`, or `isTemplate` is required, and only those update fields are accepted. For the double opt-in email (`consent.doubleOptInEmailId` from `get_tracking_settings`), new `html` or `blocks` must keep `{{DOUBLE_OPT_IN_URL}}` or nothing is saved.
set_template_localizationAcciónCreate or replace one caller-supplied localized variant of a template. The locale must be enabled for the company and cannot be its primary locale. Provide exactly one of `html` or `blocks`. By default, automatic translation on save replaces it when the original changes. Set `keepEdits` to protect it like a dashboard edit: it is marked as edited (`editedAt`) and becomes `stale` instead of being replaced.
sync_template_localizationsAcciónQueue AI translation for selected template locales. Omit `locales` to sync every enabled non-primary locale. This works even when automatic on-save localization is disabled. By default it also retranslates locales someone edited; set `skipEdited` to keep those and get them back in `skippedLocales`.
get_template_localization_coverageSolo lecturaShow how many emails that keep sending are translated into each enabled language: every email step and A/B variant in sequences that are not archived, plus enabled transactional emails (left out for marketer keys). Per language it returns total, upToDate, edited, outdated (translated before the original changed; still sends), translating, failed and missing counts. `push` gives the same counts (without failed) for push steps in sequences that are not archived or defined in code. Use it to follow progress after translate_missing_template_localizations or after a language was added in the dashboard.
translate_missing_template_localizationsAcciónQueue AI translation, in one call, for every sequence email (including sequence A/B variants) and enabled transactional email that has no translation in a language yet or whose last translation failed. It also translates sequence push steps (not archived, not defined in code) missing a language. Translations someone edited, up-to-date ones and outdated ones are left alone; use sync_template_localizations on one template to retranslate it. Omit `locales` for every enabled non-primary locale. Adding a language in the dashboard already does this for that language. Translations run in the background; follow them with get_template_localization_coverage.
delete_templateAcciónDelete a template
share_templateAcciónCreate (or fetch) the public view-in-browser link for an individual email - a transactional email, a sequence email, or a standalone template. Accepts an email/template ID or a transactional email's ID or slug; for a sequence email, pass the step's emailId from get_sequence. The hosted page renders an anonymized copy - sample contact, inert unsubscribe link, no open/click tracking - so the URL is safe to forward to anyone. Idempotent: if a link is already active, the same URL is returned with created=false rather than rotating it, so previously shared copies keep working. get_template reports the current link as shareUrl (null when none). Revoke with unshare_template. Campaigns have their own campaign-level link that follows the A/B winning variant - use share_campaign for those.
unshare_templateAcciónRevoke the email's public view-in-browser link. The shared URL returns 404 immediately, and sharing again later mints a different URL - previously distributed copies stay dead. Returns revoked=false if no link was active.
preview_cart_itemsSolo lecturaPreview cart/order data without saving or sending. Returns explicit field-mapping suggestions, ambiguous numeric price candidates, visual preset settings, and previewItems for the selected scenario. Existing itemFields are preserved. Apply chosen suggestions/preset fields using update_sequence_node or template/campaign updates; render_email accepts the previewItems at your block source in variables. Scenario data is sample-only; use real to restore the supplied items. Price candidates have unknown units: confirm cents before mapping priceCents; set priceUnit to major for major-unit prices; omitted priceUnit preserves cents. No account data is read. Requires a valid API key; no additional scopes.
get_email_block_schemaSolo lecturaField reference for Sequenzy email blocks: which fields each block type requires, which are optional, the allowed values of every enum field, and the shape of nested item arrays and nested objects. Call this before authoring blocks for a type you have not written before, or after a block write is rejected. The `blocks` parameter on create_campaign, create_sequence, create_template, and update_sequence_node accepts any of these types, but is declared as a plain object array because the union is too large to inline. Without a blockType this lists every type with its required and optional fields; with one it returns that type's full reference plus a minimal valid example and authoring notes. `group` is a persisted structural type managed by the editor and is omitted by creatableOnly; request it by name when reading existing grouped content. Note that lists are their own block type rather than a text variant: a numbered list is {"type":"list","variant":"numbered","items":[{"content":"First"}]}, and a visual numbered walkthrough is {"type":"steps","variant":"numbered","items":[{"title":"Step one","description":"What happens"},{"title":"Step two","description":"What happens next"}]}. A `text` block only accepts variants paragraph, lead, and html, and never accepts `items`. Every block also takes a `conditions` array, and each condition field accepts only its own operators - a block's own `field` and `operator` enums pool every field's operators together, so they list `is` for `tag` even though a tag condition is rejected with it. The `conditionFields` table resolves that: the operators each condition field accepts, how its `value` string is shaped, and what render_email needs before it can evaluate the condition rather than rendering it as false. It comes back automatically when you list every type and when you ask for `conditional-group`; on any other single type pass conditionFields: true, since the table is several times the size of one block's reference.
list_email_componentsSolo lecturaList the company's reusable email components newest first. Components are saved block groups that can be dropped into emails, plus the components pinned as company defaults for a slot (currently `footer`). Pass defaultsOnly to see only the pinned defaults.
get_email_componentSolo lecturaGet one email component's blocks and metadata by id, including its `version` and whether it is pinned as a company default slot.
get_default_email_componentSolo lecturaGet the company's default component for a slot - use this to read the current email footer before editing it. Returns not_found when no default is set, which means emails fall back to the footer Sequenzy generates from the company profile.
preview_default_email_componentSolo lecturaPreview a proposed default footer without saving or sending. Provide the proposed blocks and explicit application scopes. Returns affected counts, every affected/skipped email and a token for set_default_email_component. The application scopes select which categories to update: sequences, draft/scheduled campaigns, transactional templates, reusable templates, or future emails only. Customized footers are preserved unless explicitly included. Set renderPreview to get footerHtml and before/after HTML samples, one per selected category; open or save that HTML and use a browser to inspect the visual result. Use sample with an affected id/kind to render a particular email version. Previews are layout-only, without recipient personalization. A stale apply returns conflict; preview again instead of silently retrying with a new token.
set_default_email_componentAcciónCreate or replace the company's default component for a slot. This is the supported way to change the footer on every email: sequence, campaign, and AI-generated emails built from blocks clone this component when they are built, instead of the generated default footer. Raw HTML emails are the exception - block chrome cannot be injected into raw HTML, so they keep their own markup and a generated unsubscribe footer is appended at send time unless the HTML contains {{unsubscribeUrl}}. Call get_default_email_component first and edit the blocks it returns, otherwise you replace the existing footer wholesale. A default footer always keeps its unsubscribe link enabled - transactional sends hide it at render time. Omitting application keeps existing emails unchanged. To apply to existing emails, call preview_default_email_component first, then pass the same application, blocks and metadata with its token as previewToken. This atomically saves the default and updates only the previewed eligible content. It does not send emails. Customized footers are preserved by default. Conflict means the preview is stale and must be reviewed again.
create_email_componentAcciónCreate a reusable email component from a block list. Component names are unique per company. Use set_default_email_component instead when the goal is to change the footer every email gets.
| get_sync_rules | Get the company's sync rules: the automatic tag changes applied when events fire (e.g. order placed -> add a tag). New companies start with the payment and ecommerce lifecycle rules; some older ones have none. Returns the effective rules plus isDefault, which is true when the company has explicitly opted into the inherited SaaS/ecommerce platform preset. | Solo lectura |
| update_sync_rules | Replace the company's sync rules. Pass the FULL rule set (fetch with get_sync_rules first and edit it - this is not a partial update), [] to disable rules, or null to opt into the inherited SaaS/ecommerce platform preset. Each rule: { triggerEvent, actions: { addTags, removeTags }, conditions? }. Conditions support requiresTags, requiresNotTags, and purchasedProduct ({ tags?, collectionIds?, productTypes?, vendors? }) which matches products on commerce events - e.g. tag buyers of products carrying a given product tag. | Acción |
| get_email_design_system | Get the company's email design system: the visual identity every AI-generated email (campaigns and sequence steps) renders inside - design code (kicker style, title alignment, button shape, divider style, density, sanctioned opener treatments) plus the composition spine (hero-led | editorial | product-spec) that anchors email structure. The identity is stored as the company's emailDesignPrompt direction text; this tool returns the structured tokens parsed from that text (falling back to brand derivation per unstated token), the raw directionText, and isDefault (true while the identity is still purely derived from brand context). | Solo lectura |
| update_email_design_system | Adjust the company's email design system (partial update - only pass the fields to change). Affects every future AI-generated email and sequence enrichment. designCode tokens: kickerStyle (chip | letterspaced | none), titleAlignment (left | center), buttonShape (rounded | pill), dividerStyle (line | space), density (airy | compact), openerTreatments (subset of hero-image, hero-color, header-hero, editorial-masthead, title-led). compositionSpine: hero-led | editorial | product-spec. The adjustment is written into the company's emailDesignPrompt direction text (the single source of truth): the new identity's sentences are prepended and any custom prose the text carried is preserved below them. Pass reset: true to clear the direction text and return to brand-derived defaults. Free-form prose edits belong on update_company's emailDesignPrompt instead. | Acción |
| get_design_md | Get the company's design.md: its email design brief as one Markdown document. YAML frontmatter holds the exact values Sequenzy applies when rendering AI-generated emails (logo, colors, fonts.body/heading, buttons.style/radius, cornerRadius, layout tokens); prose sections hold the Voice and the Design direction the AI writer follows, plus a read-only Reference email naming the saved AI style. Values shown are the ones emails actually get. Returns markdown, version (pass it to update_design_md as expectedVersion), and the parsed designSystem. Read this before writing or redesigning emails for the company. | Solo lectura |
| update_design_md | Save an edited design.md (full document from get_design_md). Only values you changed are written: a frontmatter value left as returned is not written and keeps its current behavior, a changed radius or heading font is pinned, and a changed layout token is written into the design direction ahead of its prose. Removing a key or section leaves it unchanged; an empty Voice or Design direction section clears it (an empty direction returns to brand-derived defaults). The Reference email section is read-only. Invalid values fail with line-numbered issues; unknown keys are ignored with warnings. Pass dryRun: true to preview changes and warnings without saving. Pass expectedVersion from get_design_md; on a 409 conflict, read again and reapply your edits. | Acción |
| create_api_key | Create a new API key for a company, typically when setting up a Sequenzy integration in a project. The CLI reads the key from the SEQUENZY_API_KEY environment variable; SDK clients accept it through the apiKey option, such as apiKey: process.env.SEQUENZY_API_KEY. The key is returned only once at creation time and cannot be retrieved later. | Acción |
| request_api_key_handoff | Prepare an owner-confirmed handoff link for creating or rotating an API key. Intended for when key management fails because the active key is missing api_keys:manage; that scope fails closed on purpose, and this handoff is the supported path. Returns a URL that opens the dashboard's create-key form with the requested name and permissions already filled in. This tool creates nothing and never returns a key: the workspace owner opens the URL, reviews the form, clicks Create, and copies the new key from the browser. To rotate, pass replaceApiKeyId (or the literal string 'current' for the key you are authenticated with) and the dashboard offers to revoke the predecessor after the replacement is created. Requires only account:read. | Solo lectura |
| list_api_keys | List company-scoped API keys as non-secret metadata. Returns IDs, names, prefixes, permissions, usage timestamps, and whether each key is the active credential. It never returns a plain key or stored key hash. Use this before revoke_api_key or delete_api_key to identify the exact unused key. | Solo lectura |
| update_api_key | Rename a company-scoped API key or replace its permissions in place, without issuing a new key. Use this when a call fails with a missing-scope error: the key value stays the same, so no client has to be re-wired, and added permissions apply on the next retry. Removed permissions may remain usable for up to five minutes while API caches expire. preset and scopes REPLACE the current selection rather than merging into it - call list_api_keys first and pass the full set you want. Returns non-secret metadata only. | Acción |
| revoke_api_key | Permanently revoke a company-scoped API key by ID. The response contains non-secret metadata only. Call list_api_keys first and compare the ID, name, prefix, and isCurrent flag because revoking a key cannot be undone and may invalidate the active credential. | Acción |
| delete_api_key | Compatibility alias for revoke_api_key. Permanently delete a company-scoped API key by ID and return non-secret metadata only. Call list_api_keys first to verify the exact target. | Acción |
| list_websites | List configured sending domains with separate DNS verification, sending readiness, and SPF, DKIM, and MAIL FROM status. | Solo lectura |
| add_sending_domain | Add and configure a sending domain. A domain belongs to exactly one company and cannot be moved or removed through the API, CLI or MCP afterwards, so confirm the target company with the user before adding. Before calling verify_sending_domain, publish the DKIM record from website.dnsRecords (when dnsRecords.platformDnsRecords is present, publish those CNAMEs instead: active and spare DKIM selectors plus a recommended bounces CNAME; only required entries block verification), the bounce records described by website.bounceRecords (its cnameRecord when style is cname, otherwise spf.record and mailFrom.mxRecord; never both at the same name), and DMARC when present. Tracked links use the company tracking domain shared by every sending domain; the first domain creates it under itself (links.<domain>) and returns its CNAME as dnsRecords.trackingRecord. It is optional and never blocks verification or sending: until it verifies, links use the shared Sequenzy domain. Manage it with set_tracking_domain. | Acción |
| add_website | Compatibility alias for add_sending_domain. Adds a sending domain and returns the DNS records required for setup. The domain cannot be moved to another company afterwards; confirm the target company first. | Acción |
| check_website | Read a sending domain's separate DNS verification and sending readiness. Use verify_sending_domain to run a fresh check. dnsUpgrade.status: available (see upgrade_sending_domain_dns), in_progress (records to add), completed (oldRecords to delete) or unavailable. Receiving mail (inbox addresses and replies) is set up on receiving domains instead: see list_receiving_domains. | Solo lectura |
| verify_sending_domain | Run a fresh DNS check for a configured sending domain and return DNS verification separately from sending readiness. When readyToSend is false, read readiness.reason. Receiving mail (inbox addresses and replies) is checked on receiving domains instead: see verify_receiving_domain. | Acción |
| upgrade_sending_domain_dns | Recommended, optional one-time upgrade of a verified sending domain to the current per-domain CNAME DNS records, for better deliverability and security, without removing it; it keeps sending on current records and sending is normally not interrupted at the switch. Once started it runs to the switch. Needs website.dnsUpgrade.status available; an older SES-only domain also needs a DMARC record (400 otherwise). Publish the returned dnsUpgrade.records CNAMEs next to the current records, removing nothing; the domain switches by itself about an hour after the required ones are public (verify_sending_domain checks now, no skip). Removing dnsUpgrade.oldRecords (listed beforehand as replacedRecords) after deleteOldRecordsAfter (8 days) is optional; each is marked removed once gone from public DNS. | Acción |
| configure_sending_domain | Deprecated: tracking is company-wide; prefer set_tracking_domain. Sets <trackingPrefix>.<domain> as the company tracking domain when the company has none. Repeating the company's current tracking hostname is a no-op; a different one changes nothing and the response message says so (use set_tracking_domain to change it). Publish the returned website.tracking.cnameRecord; sending never waits for it. | Acción |
| list_integrations | List connected integrations (Stripe, Shopify, Supabase, Clerk, WooCommerce, ad platforms, and so on) with connection status, sync state, last sync time, and last sync error. Read-only: credentials, access tokens, and webhook secrets are never returned. Use this to audit which external systems feed this account and whether their syncs are healthy. | Solo lectura |
| list_sender_profiles | List sender (From) profiles and reply-to profiles for the company, including which are the account defaults and whether each sender address has verified DNS plus a ready home transport. Use this to audit sending identity before scheduling, or to pick a valid senderProfileId/replyProfileId for create_campaign or update_campaign. | Solo lectura |
| update_sender_profile | Rename one sender (From) or reply-to profile in place, without changing which profile is the account default. Use this to standardize a display name across the several identities a mailbox may carry, e.g. renaming 'Viraj from SnapCount' to 'SnapCount'. Get IDs from list_sender_profiles. Only the display name changes: the address, its sending domain, and the account-wide default From/Reply-To selection are untouched, and existing campaigns and sequences pinned to this profile pick up the new name. To change which profile is the default instead, use update_company with senderProfileId/replyProfileId. A company cannot have two sender identities with the same name on one address, so a colliding rename is rejected and reports the ID already using that name. Requires the companies:manage scope. | Acción |
| delete_sender_profile | Permanently delete one sender (From) profile. This cannot be undone. Deletion is refused when this is the company's last sender or when a live campaign, active sequence (including a step-level override), or transactional email still uses it. Draft and rejected campaigns plus the account default are reassigned to the best remaining sender when needed; read fallbackSenderProfileId in the result and review that identity before sending. Reply-to profiles are not supported by this tool. Get the profile ID from list_sender_profiles. Requires the companies:manage scope. | Acción |
| get_sending_status | Check whether company-level email sending is active, paused, or suspended, and what it takes to restore it. Relevant whenever a send, test send, or sequence step fails for a reason that is not a validation error, and whenever the account reports a bounce or complaint rate problem. Returns the pause reason and reason kind, when it was paused, the automated sender-health review state, whether the one-click resume is currently available (and if not, which gate is blocking it), the enforcement counts and thresholds for permanent bounces, temporary bounces, and complaints, and ordered remediation steps. There is no rolling window and no expiry. Enforcement uses all-time totals counted from a reset watermark, so a paused rate does NOT decay over time. Waiting does not restore sending. The rate only changes as real (non-test) sends accumulate or when a resume moves the watermark. `metricsWindow` contains the exact watermark timestamps. Requires the account:read scope, so a read-only key can diagnose a blocked account. resume_sending restores sending once the underlying problem is fixed. | Solo lectura |
| resume_sending | Request that paused company-level sending be restored after fixing the cause. This is the supported remediation path for an account paused by the permanent-bounce rate limit - it is not a bypass and it never removes suppressions. Intended for after the bad source (import, form, or integration) is actually fixed while permanent bounces remain suppressed. get_sending_status reports the current gate: for a paused workspace, resume only restores sending when `selfResume.canSelfResume` is true, which requires a hard-bounce pause reason, a cleared automated sender-health review, and no admin block. A workspace that is already active succeeds as a no-op with `resumed: false`; any other blocked state needs a support review, and the error names which one. listSanitizationConfirmed: true is required and represents the user's confirmation that the remediation is done; it is recorded on the account's audit trail as their statement that the list is clean. On success, sending is restored, the bounce watermark resets so the rate is recalculated from later sends, and the service attempts to requeue paused campaigns plus due sequence steps. The response message contains support guidance if part of that queue handoff is temporarily unavailable. Requires the companies:manage scope and owner or admin access to the company. | Acción |
| get_tracking_domain | Get the company tracking domain every sending domain uses for tracked links and opens: hostname, status, whether links use it now (active), the CNAME record to publish, and the last check. trackingDomain is null when none is set and links use the shared Sequenzy domain. | Solo lectura |
| set_tracking_domain | Set or change the company tracking domain: one hostname every sending domain uses for tracked links and opens, for example links.example.com. Any subdomain the user controls works; a bare root domain is rejected. Publish the returned trackingDomain.cnameRecord. It never blocks sending: until it verifies, links use the shared Sequenzy domain. Changing it keeps the previous hostname serving links in emails already sent while its old CNAME stays. Confirm the hostname with the user first. | Acción |
| verify_tracking_domain | Check the company tracking domain's CNAME and HTTPS certificate now instead of waiting for the automatic check. Returns verified, a message with the next step, and the tracking domain status. | Acción |
| remove_tracking_domain | Remove the company tracking domain. New emails use the shared Sequenzy tracking domain, and links and images in emails already sent through the removed domain stop working. Confirm with the user first. | Acción |
| get_frequency_cap | Get the company frequency cap: the most marketing emails (campaigns plus marketing sequence emails) one contact may receive in a rolling window. frequencyCap is null when no cap is set. Only sequences with frequencyCapEnabled respect it: their email steps are skipped for a contact already at the cap, and the contact moves to the next step. Campaigns count toward the cap but always send. Transactional emails never count and are never skipped. | Solo lectura |
| set_frequency_cap | Set or change the company frequency cap: at most maxEmails marketing emails per contact in any windowHours. Applies from the next email step onward to sequences with frequencyCapEnabled (set it with create_sequence or update_sequence). Campaigns count toward the cap but always send. Transactional emails never count and are never skipped. | Acción |
| clear_frequency_cap | Turn the company frequency cap off. Sequences keep their frequencyCapEnabled setting, but nothing is skipped until a cap is set again with set_frequency_cap. | Acción |
| get_tracking_settings | Get the company's email tracking and signup consent configuration: account-wide and Transactional API open/click defaults, unsubscribe tracking, opt-in strict bot filtering, default attribution window, automatic UTM tagging, the company tracking domain (shared by every sending domain) and its verification status, inbound reply-tracking settings, and whether double opt-in is required for new contacts. Use this to audit measurement readiness, to explain why opens or clicks may not be recorded, and to check how contacts enter the list before investigating bot or alias signups. | Solo lectura |
| update_tracking_settings | Update the company's email tracking defaults: account-wide open and click tracking, narrower open and click defaults for Transactional API sends, unsubscribe tracking, opt-in strict bot filtering, the default attribution window, and automatic UTM tagging, plus the account-wide double opt-in requirement for new contacts. Applies to emails sent afterwards; already-sent emails keep the links they were rendered with. Transactional API defaults cannot enable an account-wide setting that is off, and an individual send may opt out again. Provide at least one field. Reply tracking (inbound email, reply domain mode, reply forwarding) is on update_company. The company tracking domain is managed with set_tracking_domain, verify_tracking_domain and remove_tracking_domain. | Acción |
| get_notification_preferences | Get the account notification settings for the API key's own user in this company: whether Sequenzy emails them when a new subscriber joins, when a form is submitted, when a campaign finishes sending, and whether they get the Monday weekly report (last week's sends, engagement, new subscribers, revenue, goals, and sequence trends versus the week before; on by default for the workspace owner, off by default for invited members, only sent for weeks with more than 10 emails sent), whether they get the account health check's findings (account_health: on by default for the owner, opt-in for members; carried inside the Monday report when they get it, otherwise sent as their own email), and whether each arrives per-occurrence, as a daily summary, or weekly. Instant form-submission notifications stop after 50 per workspace per UTC day. Returns the supported modes and the platform defaults for this user's role alongside the current values. These are per-person settings - this never reads a teammate's preferences. | Solo lectura |
| update_notification_preferences | Change which account notifications Sequenzy emails the API key's own user for this company. Requires the companies:manage scope. Useful before a bulk import or a large migration, when per-signup notifications would otherwise flood the inbox, or to turn the Monday weekly report off (it is on by default only for the workspace owner; invited members set weekly_report to 'weekly' to receive it), or to opt in or out of the account health check's findings (account_health 'instant' or 'off'). Modes are 'off', 'instant' (one email per occurrence), 'daily' (one summary per day), and 'weekly' (the Monday report); 'daily' is only valid for new_subscriber, because a campaign finishes once and each form submit is its own lead, and 'weekly' is only valid for weekly_report; account_health accepts only 'off' or 'instant'. Instant form-submission notifications stop after 50 per workspace per UTC day. New-subscriber notifications already fall back to a daily summary automatically on high-volume days, and imports never trigger them at all. | Acción |
| get_integration_guide | Get code examples for integrating Sequenzy into your project. SaaS and subscription apps: when the app uses Stripe, Polar, Paddle, Dodo, Lemon Squeezy, or Creem, the native integration in the Sequenzy dashboard (Settings -> Integrations) handles payment events and customer tags automatically, so no code is needed for those. E-commerce stores: Shopify and WooCommerce stores use the native integration in the dashboard, which syncs products, orders, and customers automatically. Other e-commerce stacks (custom checkouts, CheckoutChamp, Sticky.io, headless storefronts) integrate through the 'ecommerce' use case and the Commerce API. Manual event tracking code is an opt-in use case for custom events only. Use cases: - 'transactional': Sending one-off emails (welcome, password reset, receipts) - 'subscribe_form': Adding subscribers from signup forms - 'event_tracking': Tracking CUSTOM events only (not payment events - those come from the integration) - 'ecommerce': Connecting a custom e-commerce platform via the Commerce API (sync products, push orders/checkouts, power abandoned cart + back-in-stock automations) Protected server-side API work needs an API key (create_api_key), passed to the SDK as apiKey: process.env.SEQUENZY_API_KEY. Non-Webflow static-site saved forms are the exception: list_forms/create_form/get_form_embed return public embed code, and secret keys must never be placed in browser code. Webflow-connected companies use explicitly selected native forms in Settings -> Integrations -> Webflow; custom embed code is disabled. | Solo lectura |
| get_account_health | Read the company's account health report: what to fix in its email setup. Returns eligibility (checks need more than 1,000 active subscribers and an email sent in the last 14 days; hasMinimumSubscribers and sentRecently say which is missing), settings (enabled, autonomy), nextRunAt, the latest run (status, AI summary, stats snapshot of this period vs the one before) and runInProgress, plus open and snoozed insights, most severe first. Each insight has a detector (unhandled_event, stalled_sequence, campaign_gap, stale_campaign_drafts, underperforming_sequence_email, deliverability_spike, sending_domain_failing, missing_sequence: a standard lifecycle sequence such as welcome or re-engagement the workspace does not run yet, with the suggested playbook, name, goal, trigger and steps in evidence; over_emailing: people who got 4+ marketing emails within 24 hours in the last 7 days, with overloadedSubscribers, recipients, peakPerDay, frequencyCap and the campaigns, sequences and marketing Send API email (type api, which the cap does not limit) behind it as sources in evidence, where a sequence with enrollsEveryTime and peakEnrollmentsPerPerson 3 or more re-enrolls the same people and is fixed first with update_sequence enrollmentMode one_time; neglected_subscribers: active subscribers who got no marketing email in 90 days, with neglectedSubscribers, eligibleSubscribers and estimated in evidence; unengaged_signups: contacts who joined in the last 60 days with the same consumer mailbox provider (Gmail, Yahoo, Outlook; business domains are never reported) and first list, got 2+ emails each and almost never opened or clicked while other new signups do, usually a bot filling a signup form or a bought list, with provider, providerLabel, listId, listName, emailed, sends, engaged, baselineEmailed and baselineEngaged in evidence), severity, title, detail, recommendation fixAutomationId when the agent drafted an inactive sequence to fix it, and actions: what to do about it, primary first. Each action has kind (create_sequence, open_sequence, activate_sequence, pause_sequence, create_campaign, review_campaign_drafts, open_campaign, verify_sending_domain, open_domain_settings, review_deliverability, open_frequency_cap_settings, apply_frequency_cap, review_subscribers), label, targetId, a dashboard path for screens, a confirm prompt for changes to live sending, and the api call that performs it: create_sequence -> create_account_health_fix, activate_sequence -> enable_sequence, pause_sequence -> disable_sequence, verify_sending_domain -> verify_sending_domain, open_frequency_cap_settings -> set_frequency_cap (for example maxEmails 1, windowHours 16) when evidence.frequencyCap is null, otherwise get_frequency_cap and only tighten it (the cap only holds back sequences, so when the sources are only campaigns or api it does not help; suggest spacing those out instead), apply_frequency_cap -> update_sequence with frequencyCapEnabled true on targetId (only on sequences that can wait, such as tips and nudges, not welcome or onboarding); the others open dashboard screens. Each insight also has feedback: the API key owner's own answer from submit_account_health_feedback ({rating, comment, updatedAt}), or null. Confirm with the user before set_frequency_cap or apply_frequency_cap. Confirm with the user before activate_sequence or pause_sequence. Requires analytics:read. | Solo lectura |
| update_account_health_settings | Turn the weekly account health check on or off and choose whether it drafts fixes. Pass only the fields to change. Checks never send email to subscribers or activate anything: autonomy draft_fixes only creates inactive draft sequences. Checks run every Monday at 05:00 UTC, before the Monday weekly report that carries the findings; the schedule is fixed. Who gets the findings is each person's own account_health notification preference (update_notification_preferences), not a workspace setting. Owner or admin only. Requires companies:manage. | Acción |
| run_account_health_check | Queue an account health check now and return the run. If a check is already queued or running it is returned with alreadyInProgress true instead. Manual checks are limited to one every 10 minutes and wait a day after a check that left findings open (both 429 RATE_LIMITED with retryAfterSeconds), need more than 1,000 active subscribers and an email sent in the last 14 days (409 NOT_ELIGIBLE), and never send the report email. The check usually finishes within a minute; poll get_account_health until runInProgress is false. Requires sequences:write. | Acción |
| update_account_health_insight | Act on one account health insight by id (from get_account_health): dismiss hides it for good, snooze hides it for snoozeDays (default 7, up to 90), resolve marks it fixed (it reopens if the next check still finds it), reopen brings back a snoozed or resolved one. Requires sequences:write. | Acción |
| submit_account_health_feedback | Tell the Sequenzy team whether an account health insight (id from get_account_health) was useful to the user: rating useful or not_useful, with an optional comment on why. Use it when the user says a finding or suggestion helped or missed, in their words; do not rate on their behalf. It records the API key owner's own answer and changes nothing in the workspace (to hide a finding use update_account_health_insight). One answer per person per insight: submitting again replaces the rating and keeps the earlier comment unless comment is passed (null clears it). Works for insights of any status. 404 for an unknown insight. Requires analytics:read. | Acción |
| create_account_health_fix | Build the draft sequence an account health insight asks for, from an unhandled_event or missing_sequence insight (its primary action kind is create_sequence). Creates an inactive sequence with its timing and empty, titled email steps: for unhandled_event, triggered by the event and named for the moment (for example New customer onboarding for saas.purchase); for missing_sequence, with the name, trigger (contact added, an event, or manual enrollment) and steps from the insight's evidence. Like a sequence from a template, a new draft has its emails written with AI: enrichmentStatus is processing while they are written (get_sequence reports progress), or not_queued when writing could not start (opening the sequence in the dashboard starts it; do not build again). Running it again returns the same draft with created false; steps that still have no email (for example a check's own draft nobody opened yet) are written then, with enrichmentStatus, and existing emails are never rewritten. 409 IN_PROGRESS when another build of the same insight is running; 400 for other insight types. Show it with get_sequence; enable_sequence only works once every email step has an email, and only after the user approves. Requires sequences:write. | Acción |
| get_email_ai_style | Get the company's saved AI email appearance, revisionId and canManage. Read this before replacing or clearing a style. No active style returns null; an unsupported version can still have a revisionId for recovery. | Solo lectura |
| save_email_ai_style | Save an email's appearance as the company default for future AI generation. Captures fonts, colors, spacing and block treatments, plus detected layout habits (for example dotted dividers around every button) and optional notes, without copying its content or modifying existing emails. Requires emails:write and access to the source email. Replaces only expectedStyleId; get and review the current style before replacing. All detected habits are kept unless layoutRuleIds narrows them; review style.layout.rules in the response and re-save with a subset if needed. Plain-text choices and explicit style requests take precedence during generation. | Acción |
| clear_email_ai_style | Clear the company's saved AI appearance and restore normal defaults for future generations. Existing emails stay unchanged. Requires emails:write. Get and review the current style first; a conflict must not be retried automatically. | Acción |
| disconnect_integration | Disconnect a Lemon Squeezy or SparkLoop integration and stop live ingestion and sync. Deletes only the managed provider webhook; manual webhooks remain. Local disconnection succeeds even if provider cleanup fails: inspect cleanupWarning. Repeating the call retries Lemon Squeezy cleanup; for SparkLoop, delete the leftover webhook in SparkLoop. Other providers require the dashboard. Requires integrations:manage and owner/admin access. | Acción |
| connect_integration | Connect an API-key / webhook-secret integration: polar, paddle, dodo, lemon_squeezy, whop, revenuecat, superwall, adapty, creem, chargebee, clerk, posthog, segment, mixpanel, amplitude, affonso, attio, or sparkloop. Check list_integration_capabilities first - each connectable provider lists its exact connectFields there. Credentials are validated, stored encrypted, and never returned; payment providers queue their initial revenue backfill automatically. Lemon Squeezy creates a managed signed webhook when webhookSecret is omitted, or uses the supplied secret for manual setup. RevenueCat is webhook-only (no backfill): webhookSecret is the authorization header value configured on the RevenueCat webhook, or its HMAC signing secret. Superwall is webhook-only too and needs only webhookSecret: the whsec_ signing secret of the Superwall webhook pointed at the returned webhookUrl. Mixpanel and Amplitude need only webhookSecret, a strong random value you choose: Mixpanel sends it as the Basic auth password of its webhook destination (any username, the dashboard suggests sequenzy) and syncs cohorts as mixpanel-<cohort> tags; Amplitude sends it in an X-Webhook-Secret header and streams events (product.*), user properties, and cohorts (amplitude-<cohort> tags). Sequenzy cannot show the secret again, so give it to the user once. Attio is outbound-only and has no webhookUrl. SparkLoop needs only apiKey; Sequenzy registers the SparkLoop webhook itself with a secret in its URL (webhookSecret, the account's SparkLoop-Token, is optional), and an account with several publications needs providerAccountId (the 400 error lists them in publications). Reconnecting replaces stored credentials. OAuth and app-install providers (Stripe, Shopify, Supabase, GitHub, WooCommerce, Meta) are rejected with guidance: their flows need a human in the dashboard. SECURITY: only pass credentials the user explicitly provided for this purpose; suggest the CLI (`sequenzy integrations connect`) or dashboard when the user prefers keeping secrets out of the conversation. | Acción |
| get_integration | Inspect one connected integration in depth: what the provider syncs, every event it can emit and when, every matching tag rule, which sequences trigger on those events, recent integration-specific activity, and prioritized recommendations. For Webflow, `webflowCapture` reports the exact selected native forms, retained field names, double-opt-in policy, list IDs, sequence toggle, webhook status, and schema fetch time; capture is off when its forms array is empty. `observedByAccount` and `accountLastSeenAt` are account-wide because event definitions can also be updated by another integration or the public events API; use the retained activity log for connection-specific delivery diagnosis. For Attio, an `attio_lists_unmapped` recommendation means nothing is mapped so nothing syncs, and `attio_inbound_sync_error` means changes made in Attio are not reaching Sequenzy — use get_attio_mapping then update_attio_settings; do not reconnect for that. Read-only; credentials are never returned. Get the id from list_integrations. | Solo lectura |
| list_integration_capabilities | Describe what Sequenzy's integration providers do, whether or not they are connected. For each provider: category, how it connects, what it syncs, every event it emits with the real-world moment that triggers it, the subscriber attributes it writes, supported actions, availability, and caveats. Use this to answer 'what would connecting Stripe give me?', to compare providers before recommending one, or to check whether an event a sequence needs actually exists for that provider. Coverage differs sharply between providers - only Shopify has storefront browse tracking, and only Stripe, Chargebee, and RevenueCat cover the full trial lifecycle - so check here rather than assuming. | Solo lectura |
| list_integration_suggestions | Suggest integrations to connect next, based on what Sequenzy found on the company's website during onboarding: tools on the homepage or the sign-in page it links to (Clerk, Supabase, PostHog, Segment, Mixpanel, Amplitude, Webflow, Wix, WordPress, WooCommerce) or a Shopify storefront, minus anything already connected. Payment providers are never guessed from the site: when the website lists pricing and no payment or commerce integration is connected, `paymentProviders` lists every payment provider so the user can pick theirs. `scanComplete` is false while the website is still being read, so an empty list then means 'not scanned yet', not 'nothing found'. Use list_integration_capabilities to explain what a suggested provider gives and how it connects. Read-only. | Solo lectura |
| list_integration_activity | Read recent integration webhook and sync activity, newest first. This is the log to check when an integration reports connected but contacts or events are not appearing: each row carries the action, status, event type, matched contact, and error. Payloads are sanitized at write time, so no credentials or secrets appear here. Activity is retained for 24 hours only. | Solo lectura |
| set_integration_sync_enabled | Turn a supported integration's future bulk imports and backfills on or off. THIS DOES NOT STOP AN INTEGRATION CREATING CONTACTS: disabling sync only pauses full imports and backfills, while the provider's live webhook keeps writing every new signup or customer as it happens. To change where those contacts land, use set_integration_list_targeting; to stop them arriving at all, use disconnect_integration for Lemon Squeezy or SparkLoop, or disconnect from the dashboard for other providers. Check get_integration.availableActions first; providers without enable_sync or disable_sync are dashboard-managed. Disabling keeps the connection, credentials, and live webhook delivery active, and requires any in-flight sync to finish first. It does not disconnect the integration. Idempotent: setting the current state succeeds with `changed: false`. Requires an API key with the `integrations:manage` scope, which agent-safe keys deliberately do not carry - without it, report the dashboard path (Settings -> Integrations) instead of retrying. | Acción |
| set_integration_list_targeting | Choose which lists the contacts a connected integration creates join. This is the tool for 'stop this integration adding people to my marketing lists' - set_integration_sync_enabled only pauses bulk backfills and leaves the provider's live webhook writing. Scope: this changes list membership only, and it takes effect on future provider writes. It does not stop contacts being created, does not stop their attributes syncing, does not stop sync-rule tags, and does not stop `contact_added` sequences using the default `any_contact` scope, which fire precisely because the contact joined no list. Explicit `any_list` and specific-list sequences require a matching membership and do not enroll a list-less contact. Nobody is ever removed from a list and nothing is applied retroactively. Provider behavior for existing contacts differs: Wix or Webflow submissions, Shopify customer updates, and Supabase resubscriptions can add an existing contact to the new target lists; Stripe applies targeting only when its webhook creates a subscriber. Webflow is deliberately fail-closed: `null` is rejected, so pass `[]` or explicit list IDs; form selection, field allowlists, required double opt-in, and sequence enrollment remain dashboard-only under Settings -> Integrations -> Webflow. With no selected forms, Webflow capture is off and no contacts are created. If the goal is to stop `any_contact` enrollments too, pair this tool with pause_sequence_enrollments on the affected sequences. Check get_integration first - `ingestion` reports the current targeting with list names, and `availableActions` says whether this provider honors set_list_targeting. Supported for Supabase, Stripe, Shopify, Wix, Webflow, and SparkLoop. Dodo Payments, PostHog, Polar, Paddle, and other providers without that action reject this tool; their live contacts follow defaultSubscriberListIds on get_company/update_company instead (null = every list, [] = none), which is workspace-wide. PostHog history imports are an exception and create contacts without list memberships. Idempotent: asking for the current state succeeds with `changed: false`. Requires an API key with the `integrations:manage` scope, which agent-safe keys deliberately do not carry - without it, report the dashboard path (Settings -> Integrations) instead of retrying. | Acción |
| get_integration_pixel | Check whether a Shopify store's storefront tracking pixel is installed and reporting to this account. Read live from Shopify on every call, because a merchant can remove the pixel without Sequenzy hearing about it. Check this before telling the user that browse abandonment, cart recovery, product views, collection views, or storefront search will work: without a live pixel those events never arrive, the sequences built on them never fire, and nothing anywhere reports an error. Returns `pixel.healthy`, the endpoint and signed configuration state, and `dependentEvents` - the event names that rely on this pixel. When `pixel.error` is null and `pixel.healthy` is false, those events are confirmed dark; an error means Shopify could not confirm their state. Fix a confirmed unhealthy state with activate_integration_pixel. Shopify only; other providers return a clear error. | Solo lectura |
| activate_integration_pixel | Install the Shopify storefront tracking pixel on the connected store, or repoint an existing one at this account. This is what turns on product views, cart activity, and browse abandonment - the data every browse-abandonment and cart-recovery sequence depends on. Idempotent: an already-live pixel returns `changed: false`. Events start flowing on the next storefront visit and nothing is backfilled for the period the pixel was off, so run it before, not after, building the sequence. If the store granted an older permission set, this fails with a message naming the reconnect step rather than silently doing nothing. Shopify only. | Acción |
| sync_integration | Queue a manual re-sync for an integration: customers and revenue for a payment provider (Stripe, Polar, Paddle, Dodo, Lemon Squeezy, Creem, Chargebee, Whop), the user backfill for Supabase, or the event-history import for PostHog and Segment. Use this for Supabase before a campaign that depends on names or plan attributes - the live trigger only sends rows that change after it was installed, so existing users have no attributes until a backfill runs. The Supabase sync reads the project, schema, and table already configured for the integration and cannot be pointed at a different table; if none is configured yet it fails saying so. This is also the supported way to retry a failed PostHog or Segment history import: it restarts from the beginning using the stored credentials, and already-imported events dedupe, so a re-run cannot duplicate them. Returns immediately - poll get_integration to watch `syncStatus` and `lastSyncAt`, and list_integration_activity for per-row outcomes. Fails with a conflict if a sync is already queued or running; terminal BullMQ failures release history imports for retry. Providers whose backfill is not API-triggerable (Shopify products, Affonso affiliates) return a clear error naming the dashboard. Check `availableActions` on get_integration to see whether this is supported before calling. | Acción |
| get_attio_mapping | Inspect a connected Attio integration's two-way sync: every Sequenzy list, every Attio people-list the stored token can write (id, api slug, name), the saved maps in both directions (listMap and tagMap push Sequenzy lists and tags to Attio lists; inboundListMap and inboundTagMap pull Attio lists into Sequenzy lists and tags), syncRemovals, whether company matching from work-email domains is on, and `inbound` (whether the Attio webhook is listening, and the error when it is not). Use it when Attio is connected but nothing is mapped, and call it before update_attio_settings so you have Attio list ids or slugs instead of asking the user to paste UUIDs. Reads Attio live on every call. Credentials are never returned. Attio only; other providers return a clear error. Check get_integration first — availableActions includes update_attio_settings when mapping can be saved without reconnecting. | Solo lectura |
| update_amplitude_settings | Choose which events a connected Amplitude integration records, without the webhook secret (reconnecting would replace the secret and stop deliveries until the destination is updated). syncAllEvents false plus eventAllowlist records only those types. skipInternalEvents (on for new connections) drops session, [Amplitude] autocapture, and other $ events unless allowlisted; identify always passes. Applies to later deliveries; dropped events are not replayed. Amplitude only. Requires the `integrations:manage` scope; otherwise use the dashboard (Settings -> Integrations -> Amplitude). | Acción |
| update_attio_settings | Set the two-way sync of an already-connected Attio integration, using the stored access token, so changing mappings does not require connect_integration or the Attio access token again. Sequenzy → Attio: listMap (Sequenzy list id → Attio list) and tagMap (Sequenzy tag → Attio list) upsert the person and add them to the Attio list when they join the list or get the tag. Attio → Sequenzy: inboundListMap (Attio list → Sequenzy list id) and inboundTagMap (Attio list → Sequenzy tag) create or update the subscriber and add them to the list or tag when someone is added to the Attio list, which triggers list and tag automations. Saving an inbound map makes Sequenzy register the Attio webhook itself; that needs the webhook:read-write scope on the token, and the call fails with instructions when it is missing. Each map is a full replacement: pass the complete desired map, `{}` to clear it, or omit it to leave it unchanged. syncRemovals (default false) mirrors removals in every mapped direction; off keeps sync add-only. Optionally set syncCompanyFromDomain. Provide at least one field. Idempotent: asking for the current state succeeds with changed: false. Only changes after saving sync; existing members are not backfilled. Check get_attio_mapping first for Sequenzy list ids and live Attio list ids/slugs, and list_tags for tag names. Attio only. Requires an API key with the `integrations:manage` scope, which agent-safe keys deliberately do not carry; without it, the same settings are available in the dashboard at Settings -> Integrations -> Attio -> Sync settings. | Acción |
| get_event_sample | Read the latest actual received event in the selected workspace, across all subscribers. Requires subscribers:read. Returns sample: null when none exists. Copy sample.properties into start_sequence_test_run customVariables, editing as needed. This read does not fire an event or change the selected test recipient. Unlike get_event_schema, this is recorded account data, with no age cutoff. | Solo lectura |
| get_event_schema | Read the published payload of a built-in event: a real example payload per provider, plus every property path with its type, the merge tag that resolves it, and a note wherever the example alone is ambiguous. Call this BEFORE writing {{event.*}} merge tags or event property filters - an unrecognized merge tag renders as an empty string rather than an error, so a guessed property name ships silently broken. It is also the only place that spells out the traps the sample cannot: money fields ending in `Cents` are minor units while `predictedLtv` is whole currency units, `price` is a preformatted display string that must not be parsed, `churnRisk` is a 0-95 percent and not a probability, and `orderId` is a string on Shopify but a number on WooCommerce. Omit eventName to list every documented event, or pass provider to get one provider's payload instead of all of them. Custom events are first-class and deliberately absent from this list: `documented: false` means no sample is published, never that the event name is invalid or will be rejected. Static reference data - it describes the event shape, not what this account has received. Use get_event_sample to read the latest actual received payload. | Solo lectura |
| create_template_from_example | Remix a public Sequenzy email gallery email into a new email template for this company. The new email keeps the example's exact layout and design; AI rewrites every piece of text for the company and its logo, brand color and links are swapped in. The example brand is never named and none of its copy, offers or claims are reused. Saved as an HTML template. Requires templates:write. Takes 5-20 seconds. Creates no template if generation fails (a company's first AI generation still saves its inferred email design direction); each successful call creates another template. Accepts a gallery email URL (https://sequenzy.com/email-examples/brands/{brand}/emails/{email}, or for an email in a sequence the sequence page URL ending in #email-{email}) or brand plus email slugs. For a whole sequence use create_sequence_from_example. | Acción |
| preview_template_from_example | Preview how a public Sequenzy email gallery email would read for this company without saving any email (a company's first AI generation saves its inferred email design direction, like generate_email): the same rewrite as create_template_from_example (the example's layout, every word written for the company, its logo, brand color and links swapped in), returned as subject, previewText and HTML. Pass website to write it for another site's brand instead, such as a prospect or client; it is looked up like the gallery's public preview. Requires templates:write. Takes 5-20 seconds. Call create_template_from_example to keep it as a template. Accepts a gallery email URL or brand plus email slugs. | Acción |
| preview_sequence_from_example | Preview the first 3 emails of a public Sequenzy email gallery sequence written for this company, without creating a sequence or email (a company's first AI generation saves its inferred email design direction, like generate_email). Each email keeps the example's layout with every word rewritten; returns subject, previewText, HTML and the original day for each, plus how many emails were left out. Pass website to write it for another site's brand instead. Requires sequences:write. Takes 10-30 seconds. Call create_sequence_from_example to create the draft sequence with every email. Accepts a gallery sequence URL or brand plus sequence slugs. | Acción |
| create_sequence_from_example | Clone a public Sequenzy email gallery sequence into a draft sequence for this company, with the example's trigger family (signup, abandoned cart, order placed, or inactive tag) and send timing, up to 12 emails (the first 12 unless stepNumbers picks which; each kept email keeps its original timing from the trigger). AI then writes every email in the company's brand in the background (usually 30-60 seconds); poll get_sequence until enrichmentStatus is complete. If this tool returns enrichmentStatus not_queued with a warning, the draft exists but writing did not start: open it in the dashboard and do not call this tool again. The sequence stays a draft and sends nothing until enabled. Requires sequences:write. Each call creates another sequence. Accepts a gallery sequence URL (https://sequenzy.com/email-examples/brands/{brand}/sequences/{sequence}) or brand plus sequence slugs. | Acción |
| list_email_references | List real emails from the Sequenzy email gallery to model a new email on, for the kind of email being written: campaign (promotions, newsletters, product updates), sequence (lifecycle emails plus whole sequences) or transactional. By default (scope similar) they come from the gallery brands most like the company: brands whose businesses (what each sells, and to whom) are closest in meaning to the company description, closest first (returned as brands; rankings are reused for up to 10 minutes). To rank them, the company description and any search text are sent to an embeddings service; nothing is saved. Each item has a gallery url that create_template_from_example (or create_sequence_from_example for a sequence) accepts, and emails carry an analysis (main button, length, layout, voice, structure and why it works; null until analyzed) to choose a model by. With scope all it browses the whole gallery instead; search narrows either. profile is missing (and the list empty) when the company has no description or company context yet; use scope all then. Requires templates:read. | Solo lectura |
| list_gallery_brand_requests | List this company's requests to add brands (usually competitors) to the Sequenzy email gallery, newest first. Each has a status: requested (not added yet), collecting (the brand is added and its emails are being collected), available (its emails are in the gallery; brand.url links to them and list_email_references can return them) or declined (with declineReason). Requires templates:read. | Solo lectura |
| request_gallery_brand | Ask for a brand, usually a competitor, to be added to the Sequenzy email gallery so its real emails can be used as references. Pass its website or domain. Asking again for the same domain keeps the one request (a new note replaces the old) and returns created false. If the brand is already in the gallery, the request comes back with status available. A company can keep up to 50 requests. It is not your own website. Requires templates:write. | Acción |
| remove_gallery_brand_request | Withdraw one of this company's brand requests by id (from list_gallery_brand_requests). The brand stays in the gallery if it was already added. Requires templates:write. | Acción |
| list_watched_brands | List the brands (usually competitors) this company watches, newest watch first. Each has a status: available (its emails are in the gallery; emailCount and latestEmailAt say how many and how recent), requested (not in the gallery yet), collecting (added, emails on the way) or declined (with declineReason). Use list_watched_brand_emails to read what they sent. Requires templates:read. | Solo lectura |
| watch_brand | Watch a brand, usually a competitor, by its website or domain. A brand already in the Sequenzy email gallery is watched at once (status available) and its emails can be read with list_watched_brand_emails; any other brand is requested and its emails show up once it is added and they are collected. New emails arrive a few hours after they are sent. Watching again keeps the one watch (a new note replaces the old) and returns created false. A company can keep up to 50 watches and brand requests together. It is not your own website. Requires templates:write. | Acción |
| unwatch_brand | Stop watching a brand by its watch id (from list_watched_brands). This also withdraws the brand request with the same id. The brand stays in the gallery. Requires templates:write. | Acción |
| list_watched_brand_emails | Read emails the watched brands sent, newest first, a page at a time: subject, type, subtype, sentAt, collectedAt, brand, url and a breakdown (analysis, null until analyzed). Pass domain for one brand and since for emails sent at or after a time (for example the start of this week). Emails arrive a few hours after they are sent, so when polling set since about a day before the last check and skip ids already seen. Pass nextCursor back as cursor for older emails. Each url also works as the url of create_template_from_example to remix it. Requires templates:read. | Solo lectura |
| start_subscriber_operation | Add tags to a whole filtered audience in the background. Returns a durable operation ID; poll get_subscriber_operation or list_subscriber_operations. Reuse requestKey after an uncertain response; changing inputs with the same key is rejected. Tagging needs subscribers:tag, new definitions need tags:write, and triggerAutomations needs automations:trigger. Existing synchronous bulk tag tools remain suitable for up to 500 explicit contacts. | Acción |
| list_subscriber_operations | List the company's 20 most recent background tagging operations, including progress and failures. | Solo lectura |
| get_subscriber_operation | Read background operation progress and state. A completed tagging operation may have per-contact failures; inspect failed and failures. | Solo lectura |
| cancel_subscriber_operation | Stop future work for a queued/running tagging operation. Already applied tags are retained. Tagging cancellation requires subscriber tagging access. | Acción |
| add_subscriber | Add one subscriber. Status is only applied when creating a new contact; use update_subscriber to change an existing contact's status. The subscribers:write scope covers listIds in this call. Matching sequences enroll by default (REST enrollInSequences), including tag_added when tags are set in this call, without requiring a list; pass enrollInSequences false to skip, and automations:trigger is required for enrollment. Without automations:trigger the call succeeds but starts no sequence (sequenceTriggersRan false, reason missing_scope). For multiple contacts or CRM records, use create_subscriber_import instead of looping this tool. For an email-only batch going into one list, add_subscribers_to_list is also available. | Acción |
| create_subscriber_import | Queue one CRM-grade subscriber import with up to 5,000 full contact records. Supports names, external IDs, phones, statuses, tags, lists, and custom attributes. Prefer this over repeated add_subscriber calls. Returns an import ID for get_subscriber_import. Use optInMode='confirmed' only when the contacts already gave verified email consent. Rows are imported only after a list check, usually under a minute. When the list has too many addresses that don't exist or look risky (temporary inboxes, or mostly shared addresses like info@), nothing is imported and the finished import has listQuality (status 'needs_cleaning', with issues): relay listQuality.message and suggest cleaning the list (listQuality.cleaningUrl) and importing the cleaned list. | Acción |
| get_subscriber_import | Get progress, counts, skipped reasons, and failure summaries for a queued subscriber import. Every excluded row is explained: skippedReasons sums to skippedCount and failedReasons sums to failedCount, so a finished import can be reconciled without inspecting individual contacts. If the reason counts fall short of the totals, report it with the import ID rather than guessing what the missing rows were. listQuality is null unless the list needs cleaning; then nothing was imported (rows skipped as list_needs_cleaning) and you should relay its message and cleaningUrl. An import stays running until its list check finishes. | Solo lectura |
| update_subscriber | Update an existing subscriber's profile, phone, status, attributes, or tags. Names and phones are native profile fields - set them here rather than as custom attributes. Setting status to unsubscribed suppresses the contact from all marketing, deactivates list memberships, and cancels active sequence enrollments while preserving the record and suppression history. Without automations:trigger, added tags start no sequence (sequenceTriggersRan false). | Acción |
| remove_subscriber | Unsubscribe a subscriber while preserving the record and suppression history: the contact is suppressed from all marketing, list memberships are deactivated and active sequence enrollments are cancelled. With hardDelete true the contact is permanently deleted and its history erased, which cannot be undone. Identify the contact by email or externalId. | Acción |
| get_subscriber | Get the full subscriber profile, including tags, notes, list memberships, sequence enrollments, email stats, and recent activity. For commerce customers, customAttributes include nightly predictive analytics: predictedLtv (next-12-month spend), churnRisk (percent), expectedNextOrderAt, avgDaysBetweenOrders, and predictionConfidence. | Solo lectura |
| list_subscriber_notes | List internal notes for a subscriber. Provide email or externalId to identify the subscriber. | Solo lectura |
| list_subscriber_attributes | List the custom attribute names in use across the account, each with its value type (string, number, or boolean), whether it holds a list (isArray), an example value, and how many of the sampled recent contacts carry it. Call this before add_subscriber or update_subscriber to reuse existing names and match their types, e.g. send a zip code as the string "02134" rather than a number. Names only older contacts carry are included with sampledContacts 0. mixedTypes is true when contacts hold different types for a key; valueType is then string. Always send values with leading zeros as strings, whatever valueType says. | Solo lectura |
| add_subscriber_note | Add an internal note to a subscriber. Provide email or externalId to identify the subscriber. | Acción |
| delete_subscriber_note | Delete one internal subscriber note by note ID. | Acción |
| trigger_subscriber_event | Emit a custom event for one subscriber, exactly as an integration or the public API would. This is the supported way to exercise event triggers, matching-field idempotency, branch conditions, and stop conditions end to end without waiting for real traffic. The event is recorded, sync rules apply, and matching event_received sequences enroll. Without automations:trigger the call succeeds but starts no sequence (sequenceTriggersRan false, reason missing_scope). Creates the subscriber if they do not exist. Use trigger_subscriber_events for several events on the same contact. | Acción |
| trigger_subscriber_events | Emit several custom events for one subscriber in order, exactly as an integration would: each event is recorded, sync rules apply, and matching event_received sequences enroll. Creates the subscriber if they do not exist. Events are processed independently and sequentially, so a partial failure can still leave earlier events recorded; retry with the same eventIds. Returns the subscriber and the recorded events. Without automations:trigger the call succeeds but starts no sequence (sequenceTriggersRan false, reason missing_scope). | Acción |
| import_subscriber_events | Record a bounded batch of events for many subscribers in one call, built for data pipeline and warehouse syncs. Each event carries its own email or externalId and a required source-owned eventId; externalId-only rows must identify an existing contact because email is required for creation. Contacts whose events are all more than an hour old are imported silently as history (including no double-opt-in email); any recent event makes that contact's whole group live. Retries keep one receipt and re-attempt any downstream recovery that has not finished. Accepts at most 25 events. Without automations:trigger the call succeeds but starts no sequence for live rows (sequenceTriggersRan false, reason missing_scope). | Acción |
| bulk_add_subscriber_tags | Many contacts, one or more tags: add tags to up to 500 existing subscribers in one call, identified by emails, externalIds, or subscriberIds. Never creates contacts. Built for reconciling historical or derived tags: subscribers that do not exist are reported in notFound instead of being created, and tag automations do NOT run unless triggerAutomations is true. Returns the normalized tags and counts of requested, matched, updated, unchanged and failed subscribers, with notFound identifiers and up to 50 failures. Needs the subscribers:tag scope; a tag name that does not exist yet also needs tags:write. The agent_safe key preset includes both. Use add_subscriber or update_subscriber for single-contact changes. | Acción |
| bulk_remove_subscriber_tags | Many contacts, one or more tags: remove tags from up to 500 existing subscribers in one call, identified by emails, externalIds, or subscriberIds. Never creates contacts: subscribers that do not exist are reported in notFound. Needs the subscribers:tag scope; the broader subscribers:write scope also satisfies it. Use this to roll back or reconcile a derived-tag backfill. Returns the normalized tags and counts of requested, matched, updated, unchanged and failed subscribers, with notFound identifiers and up to 50 failures. | Acción |
| search_subscribers | Search subscribers by free-text query, tags, list, segment, custom attribute, or opt-out date. Each subscriber carries unsubscribedAt, the date the contact opted out - use it rather than updatedAt, which any later tag or attribute write overwrites. If you omit limit, the tool fetches all pages and returns every match in one call - that is the fastest way to enumerate an audience. To walk a large audience in chunks instead, pass limit and then follow pagination.nextCursor (or pagination.nextOffset) until pagination.hasMore is false. Default searches are newest first; searches with attribute use stable subscriber-ID ascending order. Returns the matching subscriber records, the count returned, and pagination with total, hasMore, nextCursor and nextOffset. | Solo lectura |
| list_accounts | List your customers' organizations (B2B accounts such as companies, workspaces or teams) in the workspace. The dashboard shows them under Subscribers > Organizations; these tools keep the account names. Not your Sequenzy account, workspace or login: use get_account for that. Search matches name, external ID and domain. Organizations group contacts; their attributes reach every member as `organization.<name>` for segment attribute filters and sequence conditions, and as `{{organization.<name>}}` merge tags (`account.*` is an older alias). | Solo lectura |
| get_account_by_external_id | Get one customer organization (Subscribers > Organizations in the dashboard) by its external ID, including up to 100 member contacts with their roles (owner, admin, member). Not your Sequenzy account: use get_account for that. | Solo lectura |
| upsert_account | Create or update a customer organization (Subscribers > Organizations in the dashboard) by external ID. Attributes merge into the stored attributes (a null value deletes a key; set replaceAttributes to replace them all). Optional members are added in the same call. Use this to model companies, workspaces or teams so segments can filter on `organization.plan` and emails can use `{{organization.name}}` (`account.*` still works). The first organization turns Organizations on for the workspace; it fails with a 409 error when contacts already use a custom attribute named `account` or `account.*`, which must be renamed first. | Acción |
| delete_account | Delete one customer organization (a B2B account your contacts belong to, shown under Subscribers > Organizations) and its memberships. Contacts are kept; their `organization.*` attributes are cleared. It never deletes the user's Sequenzy account, workspace (company), login or API keys. If the user asks to delete their own Sequenzy account or workspace, do not call this tool: that is only possible in the Sequenzy dashboard. Permanent; confirm the organization's externalId with the user first. | Acción |
| list_account_members | List the contacts that belong to a customer organization with their roles. Owners first, then admins, then members. | Solo lectura |
| add_account_member | Add a contact to a customer organization (creating the contact when the email is new). Re-adding an existing member with a role updates the role; without a role keeps the current one. | Acción |
| remove_account_member | Remove a contact from a customer organization. The contact is kept; its `organization.*` attributes are cleared or replaced by another organization it belongs to. | Acción |
| trigger_account_event | Record an event on a customer organization (for example trial_ending, plan_changed, seat_limit_reached) and deliver it as a contact event to the chosen recipients: owners (default; falls back to admins when there is no owner), admins (includes owners), all members, or none (timeline only). Each delivery runs sync rules and sequence triggers with `event.organization.*` properties (also available as `event.account.*`). To start a sequence once per organization rather than once per recipient, create it with enrollmentMode matching_field and enrollmentFieldPath organization.externalId. Pass eventId to make retries idempotent. | Acción |
| list_account_events | List recent events on a customer organization's timeline, newest first. | Solo lectura |
| list_account_suggestions | Suggest accounts from work email domains that several contacts share (for example 7 contacts at acme.com). Personal and disposable providers (gmail.com, outlook.com, yopmail.com, ...), the workspace's own sending domains, domains where any contact already belongs to an account, and domains an account already uses are skipped. Each suggestion includes up to 10 sampleEmails for review. Read-only; show the list to the user and accept only the domains they confirm with accept_account_suggestions. | Solo lectura |
| accept_account_suggestions | Create accounts from suggested email domains. Each domain becomes an account (externalId and domain set to the domain, name derived from it) with the domain's contacts that belong to no account as members. When one account already uses the domain, contacts are added to it; with several, the domain is skipped (multiple_accounts). A domain whose contacts already partly belong to another account is skipped (already_in_account); domains without eligible contacts are skipped (no_contacts). Up to 25 domains per call; personal, disposable and the workspace's own sending domains are rejected. Keeps existing roles, runs no sync rules or segment-entered sequences, and a retry does not duplicate accounts or members; truncated means contacts were left out because an account holds at most 5,000 members. Fails with a 409 error while contacts use a custom attribute named `account` or `account.*`. | Acción |
| detect_account_organization_ids | Find the organization ID the app already sends, so accounts can be created from existing data without code. Looks at event properties from the last 180 days and contact attributes named like workspaceId, organization_id, team_id or account_id. Each candidate has source (event or attribute), propertyKey, nameKey (a sibling name property or null), eventCount (events for event, contacts for attribute), accountCount (distinct organizations) and eventNames. Read-only; next, preview a candidate with preview_accounts_from_organization_id and show it to the user. | Solo lectura |
| preview_accounts_from_organization_id | Preview the biggest accounts create_accounts_from_organization_id would make, before anything is written. Each organization has externalId (its ID), name (from nameKey, otherwise from the work email domain most of its contacts share, null when neither exists), domain, contactCount and up to 3 sampleEmails. Read-only; always show this preview to the user and get confirmation before creating accounts. | Solo lectura |
| create_accounts_from_organization_id | Start a background job that creates one account per distinct propertyKey value and adds each contact seen with it as a member. Accounts are named from nameKey, otherwise after the work email domain most of their contacts share; existing names, domains and roles are kept. No emails are sent, no sync rules run and no segment-entered sequences start. Returns jobId, alreadyRunning (a run for the same key was already queued or running, so none was started) and job; check progress with get_account_organization_id_job. Reruns update the same accounts instead of duplicating them. Only call after the user confirmed the preview. Fails with a 409 error while contacts use a custom attribute named `account` or `account.*`. | Acción |
| get_account_organization_id_job | Check a create_accounts_from_organization_id run. state is queued, running, completed or failed. A completed job has result with accountsCreated, accountsUpdated, membershipsCreated, membershipsUpdated, skipped and truncated (history was larger than one run covers; start it again to continue). A failed job has error. Finished jobs expire after a while (404); the accounts stay. | Solo lectura |
| list_recipient_suppressions | List the recipients this company cannot reach: protected global invalid-recipient suppressions for addresses associated with the company, protected workspace hard bounces without conclusive invalid-inbox evidence, removable workspace soft-bounce escalations, and protected spam complaints. Each row includes a stable suppressionType, and the response echoes the applied sortBy/sortOrder. Sort by `status` to surface the suppressions you can actually remove. Start here when a customer reports missing email; use get_recipient_suppression for one exact address. | Solo lectura |
| get_recipient_suppression | Check whether one recipient is suppressed by Sequenzy's bounce/complaint safeguards. Suppression comes solely from Sequenzy's own bounce records; the email provider's suppression list is not consulted. This exact-address lookup never exposes other recipients. | Solo lectura |
| remove_recipient_suppression | Remove a workspace-scoped soft-bounce escalation for an exact recipient associated with the company, then reactivate bounced company subscribers. Global invalid-recipient, workspace hard-bounce, complaint, and unsubscribe suppressions are protected. | Acción |
| list_products | List synced products (Stripe, Shopify, WooCommerce) including any attached digital delivery file. Useful before attaching a file or building a purchase sequence for a specific product. For Stripe products, variants are the product's active prices (variantId = Stripe price ID), so an exact price can be targeted even when it is not the default. Returns one page: check pagination.total and pagination.hasMore, and page with offset to read the whole catalog. | Solo lectura |
| upsert_products | Create or update products in the catalog (Commerce API, keyed by your productId, stored under the api provider). Use this to add products that are not synced from Stripe/Shopify/WooCommerce, then attach a deliverable file with attach_product_file and build a purchase sequence for them. | Acción |
| delete_product | Delete a product previously pushed via upsert_products (Commerce API products only), identified by your productId. | Acción |
| attach_product_file | Attach a distributable file to a product, either by public URL or by uploading a local file (filePath, local MCP server only). After a purchase of the product, sequence emails can link to it with {{event.download.url}} and {{event.download.name}}. | Acción |
| remove_product_file | Remove the attached distributable file from a product. | Acción |
| sync_products | Queue a sync of the Stripe product catalog into the products list. Requires an active Stripe integration with bulk sync enabled. Pass integrationId when more than one Stripe account is connected; otherwise the most recently connected enabled account is used. | Acción |
| get_push_settings | Get push notification readiness: whether web push is enabled (with its VAPID public key), whether APNs (iOS) and Firebase (Android) credentials are connected, the last credential error for each, and active device counts per platform. Check this before creating push campaigns or push sequence steps. | Solo lectura |
| update_web_push_settings | Enable or disable web push, or set the default notification icon. The first enable generates the workspace's VAPID keys; disabling keeps them so existing browser subscriptions stay valid. The site must serve /sequenzy-sw.js (importScripts of the hosted worker) and call sequenzy.subscribeToPush() from a click. | Acción |
| set_apns_credentials | Connect iOS push by saving an APNs auth key (.p8). The key is encrypted at rest and never returned. Requires the integrations:manage scope and a role that can edit workspace settings (marketer keys cannot). | Acción |
| remove_apns_credentials | Remove APNs credentials. iOS devices stop receiving push until a key is saved again; registered devices are kept. | Acción |
| set_fcm_credentials | Connect Android push by saving a Firebase service account key (the JSON file from Firebase project settings > Service accounts). Encrypted at rest and never returned. | Acción |
| remove_fcm_credentials | Remove Firebase credentials. Android devices stop receiving push until a key is uploaded again; registered devices are kept. | Acción |
| list_push_devices | List registered push devices (browsers and app installs) in a stable order; pass nextCursor to page. Filter by contact (email or subscriberId), platform or status. Tokens are never returned in full. | Solo lectura |
| register_push_device | Register (or refresh) a push device for a contact, the way an app backend would after the app receives its token. Re-registering a token reactivates it and moves it to the given contact. Unknown emails become contacts quietly (no lists, no contact_added sequences). | Acción |
| remove_push_device | Unsubscribe a push device and unlink it from its contact so it stops receiving push. A browser or app that still allows notifications can register again (browsers come back unlinked). | Acción |
| send_test_push | Send a real test push to one device (deviceId) or to every active device of a contact (email or subscriberId). Provide title and/or body. Limited to 200 per company per rolling 24 hours; excluded from stats. | Acción |
| list_push_campaigns | List push campaigns, newest first. | Solo lectura |
| get_push_campaign | Get a push campaign's content, audience, status and (after sending) delivery stats. | Solo lectura |
| create_push_campaign | Create a push campaign draft. Content and audience can be set now or later with update_push_campaign; send with send_push_campaign. | Acción |
| update_push_campaign | Update a draft or scheduled push campaign. Only the fields you pass change. A scheduled campaign keeps its schedule; timing fields can only change on drafts (reschedule with send_push_campaign instead). | Acción |
| estimate_push_campaign_recipients | Count how many contacts in an audience have an active push device (eligibleCount) versus the whole audience (audienceCount). | Solo lectura |
| send_push_campaign | Send a push campaign now, or schedule it with scheduledAt (future ISO 8601). Optionally deliver in each contact's local time (sendInRecipientTimezone + scheduledTimezone) or at each contact's optimal hour (sendTimeOptimization); omitted timing fields keep the draft's saved values. Delivers real notifications to every contact in the audience with an active device on a set-up platform. Confirm with the user before calling. | Acción |
| unschedule_push_campaign | Move a scheduled push campaign back to draft. | Acción |
| cancel_push_campaign | Stop a scheduled or sending push campaign. Unsent notifications are skipped; delivered ones cannot be recalled. | Acción |
| duplicate_push_campaign | Copy a push campaign (content and audience) into a new draft. | Acción |
| get_push_campaign_stats | Get push campaign delivery stats: sent, displayed (device-reported), clicked, failed and skipped contacts, devices targeted/accepted, and skip/failure reasons. | Solo lectura |
| translate_push_campaign | Translate a draft or scheduled push campaign's title and message with AI into the workspace's languages and save the versions. Contacts then get the version for their language attribute or device language. Versions a person wrote are kept unless overwrite is true. Returns the campaign plus translated, skipped and failed languages. | Acción |
| translate_push_content | Translate a push title and message with AI into the workspace's languages without saving anything. Use the returned localizations for a sequence push step (update_sequence_node changes.localizations, or localizations on a push step in create_sequence/insert_sequence_step) or a push campaign. | Solo lectura |
| get_push_import_preflight | Check push settings before importing devices: warns when APNs or Firebase is not set up, or APNs is on sandbox. Imported tokens only work with the same bundle ID, Apple team and Firebase project the app used with the previous provider. Show these warnings to the user before starting an import. | Solo lectura |
| list_push_imports | List recent push imports with status, progress and counts (imported, matched, unsubscribed, skipped, invalid, contacts created). | Solo lectura |
| get_push_import | Get one push import: status (draft, queued, exporting, importing, completed, failed, canceled), progress, counts, skip/invalid reasons, warnings and whether it can be resumed. Poll this while it runs. | Solo lectura |
| create_push_import | Create a draft push import and return its warnings; nothing is read until start_push_import. Requirements by source: onesignal needs apiKey (App API key) and onesignalAppId; braze needs apiKey (REST key with users.export.segment), brazeRestEndpoint and brazeSegmentId; customer_io needs apiKey (App API key), customerIoSegmentId and optionally customerIoRegion; csv takes the file as csv (token and platform columns, plus external_id or email; at most 5 MB, send more with add_push_import_rows and set csvParts to the total part count so a partial upload can never start). Opted-in devices are linked to contacts by external ID or email (unmatched devices get a new contact, quietly: no lists, no sequences; push-only with no email or phone when the device has only an external ID); opted-out devices are kept as unsubscribed. | Acción |
| add_push_import_rows | Add another part of a CSV file to a draft csv import (at most 5 MB per part; every part starts with the header row). Parts are imported in the order they are added. | Acción |
| start_push_import | Start a draft push import, or resume a failed or stalled one from where it stopped. Runs in the background; poll get_push_import. One import runs per workspace at a time. | Acción |
| cancel_push_import | Stop a push import, or discard a draft or failed one. Devices already imported stay; staged rows and the stored source key are deleted. Cannot be undone. | Acción |
| list_live_activities | List iOS Live Activity and Android Live Update tokens (push-to-start tokens per device and activity type, update tokens per running activity) with a per-type summary of devices that can start activities and running activities. Each token has platform ios or android. Tokens are masked. | Solo lectura |
| start_live_activity | Start a Live Activity remotely on the devices that registered a push-to-start token for activityType: iOS 17.2+ through APNs, and Android Live Updates through FCM on apps that observe the same type name with the Sequenzy Android SDK (promoted on Android 16, an ongoing notification before). Target one contact (email or subscriberId), or every contact in a segment, list or tag (exactly one of segmentId, listId, tag; up to 100,000 devices, needs campaigns:send), then update and end it with broadcast: true. Requires an alert. Returns tokenCount 0 without sending when no device can start it; contactCount and platforms count contacts and devices per platform. Repeating the call starts another copy. | Acción |
| update_live_activity | Update every running copy of a contact's Live Activity (one per device, iOS and Android Live Updates) with new content state. Requires email or subscriberId, or broadcast: true to update every signed-in device running the activityId (for activities started for a segment, list or tag). Add an alert to light the screen. Returns tokenCount 0 when no running copy has reported an update token. | Acción |
| end_live_activity | End a contact's Live Activity on every device running it (iOS and Android Live Updates), with its final content state and optionally a dismissal date. Requires email or subscriberId, or broadcast: true to end it on every signed-in device running the activityId. Without dismissalDate iOS keeps it on the Lock Screen and Android in the shade for up to four hours. | Acción |
| list_in_app_messages | List a contact's in-app messages and content cards (created by type "in_app" sequence steps). status "open" (default) is what their app should show now; "all" is history. Identify the contact by email, externalId or subscriberId. | Solo lectura |
| record_in_app_event | Record that the app displayed, clicked, dismissed or read an in-app item. Repeats change nothing except clickCount. Clicks, dismissals and reads also count as a display. Messages leave the open feed once displayed, clicked or dismissed; cards once dismissed. | Acción |
| delete_in_app_message | Delete an in-app message or card so the app stops showing it. | Acción |
| get_in_app_step_stats | Created, displayed, clicked and dismissed counts for one in-app sequence step, plus how many contacts allowed (pushPermissionGranted) or declined (pushPermissionDenied) notifications from a push primer. | Solo lectura |
| list_in_app_templates | Starting points for in-app sequence steps (trial ending, welcome, new feature, milestone, a push primer that asks for notification permission, and more). Each has a complete step config: pass it as a type "in_app" step to create_sequence or insert_sequence_step after adding the urls its link buttons open. | Solo lectura |
| validate_in_app_html | Check a custom HTML in-app message document before using it as a format "html" in-app step. Returns its size against the 100 KB limit, Braze Liquid conversions (e.g. {{${email_address}}} becomes {{email}}, {{${braze_id}}} becomes {{contactId}}), script hosts a website CSP must allow, whether it closes itself, warnings and blocking errors. Nothing is saved. | Solo lectura |
| write_in_app_copy | Write in-app message copy with AI from a short brief: eyebrow, title, message and button labels sized for the layout. Returns text only; nothing is saved. Button links are never invented, so add urls before using the copy in a step. | Solo lectura |
| list_in_app_campaigns | List in-app campaigns, newest first. Each has a stored status and a state: draft, scheduled, live, ended (past endsAt) or stopped. | Solo lectura |
| get_in_app_campaign | Get an in-app campaign's message, audience, priority, schedule, end time and state, plus reach and engagement stats once it has launched. | Solo lectura |
| create_in_app_campaign | Create an in-app campaign draft: one in-app message or content card for an audience, shown inside the customer's app. Set the message and audience now or later with update_in_app_campaign; launch with send_in_app_campaign. | Acción |
| update_in_app_campaign | Update a draft or scheduled in-app campaign. Only the fields you pass change; message fields merge over the stored message. Live campaigns can't be edited: stop and duplicate them instead. | Acción |
| estimate_in_app_campaign_recipients | Count contacts who match an audience now (audienceCount) and how many are known to have your app connected (appUserCount, a lower bound). Everyone in the audience gets the campaign the next time their app fetches the feed, including contacts who join while it is live. If appUserCount is 0, say Sequenzy hasn't seen their app connect yet; if in-app isn't set up, point to Settings > In-app. | Solo lectura |
| send_in_app_campaign | Launch an in-app campaign now, or schedule it with scheduledAt (future ISO 8601). While live, every contact in the audience sees the message the next time their app fetches its feed. Confirm with the user before calling. | Acción |
| unschedule_in_app_campaign | Move a scheduled in-app campaign back to draft. | Acción |
| stop_in_app_campaign | Stop a scheduled or live in-app campaign. Nobody new gets it and it leaves every feed that still has it, including seen cards; stats are kept. Stopping is final (duplicate to run it again); stopping again succeeds. | Acción |
| duplicate_in_app_campaign | Copy an in-app campaign (message, priority, audience and a future end time) into a new draft. | Acción |
| get_in_app_campaign_stats | Get in-app campaign results: delivered (contacts whose app received it), displayed, clicked, totalClicks, dismissed, push permission results for primer buttons, and first clicks per custom HTML button, plus the audience size at launch. | Solo lectura |
| list_in_app_placements | List the workspace's in-app placements: named slots in the customer's app or website that show in-app content inline (key, name, description, size hints) with all-time counts (created, live, displayed, clicked, dismissed). Fill one from a sequence step with type "in_app", kind "placement" and placement set to its key. | Solo lectura |
| get_in_app_placement | Get one in-app placement by key, with its all-time counts: items created, live (open now), and contacts who displayed, clicked or dismissed their item. | Solo lectura |
| create_in_app_placement | Create an in-app placement: a slot in the customer's app that shows in-app content inline, such as a home screen hero. The key is what app code passes to the SDK and never changes. Up to 50 per workspace. | Acción |
| update_in_app_placement | Rename an in-app placement or change its description or size hints. The key cannot change. Provide at least one field. | Acción |
| delete_in_app_placement | Delete an in-app placement. Its slots collapse in the app on the next refresh, and in-app steps that target it skip contacts until a placement with the same key exists again. Items already created stay in contact history. | Acción |
| get_in_app_placement_content | What each placement shows one contact now: for each key, the contact's highest-priority open placement item (ties: newest), or null when the slot collapses. Identify the contact by email, externalId or subscriberId. Reading marks nothing displayed. | Solo lectura |
| list_tags | List every tag defined in the company, with each tag's ID, normalized name and color. Use it to reuse existing tag names before tagging contacts, or to find a tagId. | Solo lectura |
| create_tag | Create a new tag definition. The name is normalized to lowercase with hyphens (e.g. 'VIP Customer' becomes 'vip-customer'). Color defaults to gray. Returns the created tag with its ID, normalized name and color. Creating a tag does not apply it to any contact. | Acción |
| update_tag | Update a tag's color. System tags cannot be updated. | Acción |
| delete_tag | Permanently delete a tag and remove it from all subscribers. This cannot be undone. System tags and tags used by sequences cannot be deleted. | Acción |
| list_lists | List all subscriber lists. Each list includes subscriberCount (current members of any status) and activeSubscriberCount (current members with status=active). Members who unsubscribed from the list are not counted. Use these fields for list audience size instead of paging search_subscribers. | Solo lectura |
| create_list | Create a new subscriber list. Lists are public by default: public lists appear to subscribers in the email preference center, where they can join or leave them. Set isPrivate to keep the list internal. | Acción |
| update_list | Update a subscriber list's name, description, or privacy. Only the provided fields are changed. A public list's name is shown to subscribers in the email preference center, and making a list public lets subscribers join it there. | Acción |
| delete_list | Permanently delete a subscriber list and remove all of its memberships. Subscribers themselves are kept. This cannot be undone. | Acción |
| add_subscribers_to_list | Bulk add subscribers to a subscriber list from an email array. Existing subscribers are added to the list without a per-subscriber update call; addresses not yet in the account are created as new subscribers, using optInMode for their consent. Returns the list plus added and skipped counts. Get listId from list_lists or create_list. | Acción |
| remove_subscribers_from_list | Remove subscribers from a list by email address. Maximum 500 emails per call. Subscribers stay in the account; only the list membership is removed. Returns the removed count plus a notFound array of emails that did not match a subscriber. | Acción |
| list_segments | List every saved segment in the company with its ID and name. Use the IDs with get_segment_count, search_subscribers segmentId, or campaign targeting. | Solo lectura |
| create_segment | Create a new segment from explicit filter rules. Use `filters` plus `filterJoinOperator` for flat legacy rules, or `root` for nested AND/OR groups such as `{ "kind": "group", "joinOperator": "and", "children": [{ "kind": "filter", "field": "attribute", "operator": "gte", "value": "mrr:50" }, { "kind": "group", "joinOperator": "or", "children": [{ "kind": "filter", "field": "tag", "operator": "contains", "value": "vip" }, { "kind": "filter", "field": "event", "operator": "is_not", "value": "saas.purchase:30d" }] }] }`. AND of two or more positive attribute filters on the same array path (e.g. `history_events[].eventvenue_id` and `history_events[].showing_date`) requires one array element to satisfy every condition — a 2010 visit at that venue plus an unrelated recent booking elsewhere does not match. Negative filters (`is_not`, `not_contains`, `is_empty`) are exempt: they assert no element matches and combine independently. Supports `event` and `segment` fields, Stripe product purchase/current/trial/date filters, and campaign- or delivery-policy-scoped engagement filters such as `emailOpened is marketing:all`. | Acción |
| update_segment | Update a segment's name and/or filter rules. Use the same `filters` plus `filterJoinOperator` or nested `root` shapes as create_segment; providing `filters` or `root` replaces the segment's existing rules. | Acción |
| delete_segment | Permanently delete a segment. This cannot be undone. Subscribers are not affected. Deleting is never blocked when the segment is in use: the result lists `affectedSequences` (with how each uses it), `affectedCampaigns`, `affectedSegments` and `affectedAudienceSyncs`, which from then on treat every contact as not in the segment. Requires the segments:delete scope; segments:write is not enough. | Acción |
| get_segment_count | Count the subscribers who currently match a saved segment's rules. Returns count and the segmentId that was counted; it does not list the subscribers (use search_subscribers with segmentId for that). | Solo lectura |
| list_audience_syncs | List Meta custom audience syncs (segment -> ad audience mappings) with their schedule and last sync status. Requires the Meta Ads integration to be connected in the dashboard. | Solo lectura |
| list_ad_accounts | List the Meta ad accounts available through the connected Meta Ads integration. Use the returned account id as `adAccountId` when creating an audience sync. | Solo lectura |
| create_audience_sync | Push a segment to a Meta custom audience and keep it synced on a schedule. Provide `segmentId` for an existing segment OR `predefinedSegmentId` for a ready-made template (for example "zero-ltv", "no-purchase-1y", "recent-buyers", "high-spenders-ecom", "non-buyers", "engaged") - the template segment is created automatically on first use. The first upload runs immediately. Note: audiences are add-only; subscribers who later leave the segment stay in the Meta audience. | Acción |
| update_audience_sync | Update an audience sync's frequency or pause/resume it (`isActive`). | Acción |
| delete_audience_sync | Remove an audience sync mapping. The Meta audience itself is kept (running ads are not disrupted) - only future syncs stop. | Acción |
| sync_audience_now | Trigger an immediate upload of the segment's subscribers to its Meta custom audience, outside the regular schedule. | Acción |
| list_templates | List the company's email bodies newest first, including per-locale localization sync status and each body's `emailPreset` Style > Format (branded or minimal; null for standalone raw HTML). Results include standalone templates plus the bodies behind campaigns and transactional emails. Returns 50 per call by default (100 max); page with offset while pagination.hasMore is true, and read pagination.total for the full count. You can also pass a campaign's emailId to get_template to fetch one body directly. Bodies are kept when their campaign or transactional email is deleted. | Solo lectura |
| get_template | Get a template's details, content, and all localized variants with sync status. `emailPreset` reports the per-email Style > Format (branded or minimal; null when the whole email is standalone raw HTML) the same way get_sequence does per step. | Solo lectura |
| create_template | Create a new email template. `prompt` generates net-new email content from natural language. `blocks` is for finished caller-supplied Sequenzy content and `html` for supplied or explicitly requested preserved HTML. Creating a standalone copy of a saved design and AI rewriting within its layout are currently dashboard-only workflows; this tool has no source-design operation, and `prompt` does not preserve an existing layout. For a campaign copy, use create_campaign with templateId. | Acción |
| update_template | Update an existing template in place: the supplied fields replace the saved ones (labels replace the whole label set). Campaigns and sequence steps that started from this design keep their own copies; automatically translated variants are refreshed from the new original unless marked keepEdits (see set_template_localization). At least one of `name`, `subject`, `previewText`, `html`, `blocks`, `labels`, or `isTemplate` is required, and only those update fields are accepted. For the double opt-in email (`consent.doubleOptInEmailId` from `get_tracking_settings`), new `html` or `blocks` must keep `{{DOUBLE_OPT_IN_URL}}` or nothing is saved. | Acción |
| set_template_localization | Create or replace one caller-supplied localized variant of a template. The locale must be enabled for the company and cannot be its primary locale. Provide exactly one of `html` or `blocks`. By default, automatic translation on save replaces it when the original changes. Set `keepEdits` to protect it like a dashboard edit: it is marked as edited (`editedAt`) and becomes `stale` instead of being replaced. | Acción |
| sync_template_localizations | Queue AI translation for selected template locales. Omit `locales` to sync every enabled non-primary locale. This works even when automatic on-save localization is disabled. By default it also retranslates locales someone edited; set `skipEdited` to keep those and get them back in `skippedLocales`. | Acción |
| get_template_localization_coverage | Show how many emails that keep sending are translated into each enabled language: every email step and A/B variant in sequences that are not archived, plus enabled transactional emails (left out for marketer keys). Per language it returns total, upToDate, edited, outdated (translated before the original changed; still sends), translating, failed and missing counts. `push` gives the same counts (without failed) for push steps in sequences that are not archived or defined in code. Use it to follow progress after translate_missing_template_localizations or after a language was added in the dashboard. | Solo lectura |
| translate_missing_template_localizations | Queue AI translation, in one call, for every sequence email (including sequence A/B variants) and enabled transactional email that has no translation in a language yet or whose last translation failed. It also translates sequence push steps (not archived, not defined in code) missing a language. Translations someone edited, up-to-date ones and outdated ones are left alone; use sync_template_localizations on one template to retranslate it. Omit `locales` for every enabled non-primary locale. Adding a language in the dashboard already does this for that language. Translations run in the background; follow them with get_template_localization_coverage. | Acción |
| delete_template | Delete a template | Acción |
| share_template | Create (or fetch) the public view-in-browser link for an individual email - a transactional email, a sequence email, or a standalone template. Accepts an email/template ID or a transactional email's ID or slug; for a sequence email, pass the step's emailId from get_sequence. The hosted page renders an anonymized copy - sample contact, inert unsubscribe link, no open/click tracking - so the URL is safe to forward to anyone. Idempotent: if a link is already active, the same URL is returned with created=false rather than rotating it, so previously shared copies keep working. get_template reports the current link as shareUrl (null when none). Revoke with unshare_template. Campaigns have their own campaign-level link that follows the A/B winning variant - use share_campaign for those. | Acción |
| unshare_template | Revoke the email's public view-in-browser link. The shared URL returns 404 immediately, and sharing again later mints a different URL - previously distributed copies stay dead. Returns revoked=false if no link was active. | Acción |
| preview_cart_items | Preview cart/order data without saving or sending. Returns explicit field-mapping suggestions, ambiguous numeric price candidates, visual preset settings, and previewItems for the selected scenario. Existing itemFields are preserved. Apply chosen suggestions/preset fields using update_sequence_node or template/campaign updates; render_email accepts the previewItems at your block source in variables. Scenario data is sample-only; use real to restore the supplied items. Price candidates have unknown units: confirm cents before mapping priceCents; set priceUnit to major for major-unit prices; omitted priceUnit preserves cents. No account data is read. Requires a valid API key; no additional scopes. | Solo lectura |
| get_email_block_schema | Field reference for Sequenzy email blocks: which fields each block type requires, which are optional, the allowed values of every enum field, and the shape of nested item arrays and nested objects. Call this before authoring blocks for a type you have not written before, or after a block write is rejected. The `blocks` parameter on create_campaign, create_sequence, create_template, and update_sequence_node accepts any of these types, but is declared as a plain object array because the union is too large to inline. Without a blockType this lists every type with its required and optional fields; with one it returns that type's full reference plus a minimal valid example and authoring notes. `group` is a persisted structural type managed by the editor and is omitted by creatableOnly; request it by name when reading existing grouped content. Note that lists are their own block type rather than a text variant: a numbered list is {"type":"list","variant":"numbered","items":[{"content":"First"}]}, and a visual numbered walkthrough is {"type":"steps","variant":"numbered","items":[{"title":"Step one","description":"What happens"},{"title":"Step two","description":"What happens next"}]}. A `text` block only accepts variants paragraph, lead, and html, and never accepts `items`. Every block also takes a `conditions` array, and each condition field accepts only its own operators - a block's own `field` and `operator` enums pool every field's operators together, so they list `is` for `tag` even though a tag condition is rejected with it. The `conditionFields` table resolves that: the operators each condition field accepts, how its `value` string is shaped, and what render_email needs before it can evaluate the condition rather than rendering it as false. It comes back automatically when you list every type and when you ask for `conditional-group`; on any other single type pass conditionFields: true, since the table is several times the size of one block's reference. | Solo lectura |
| list_email_components | List the company's reusable email components newest first. Components are saved block groups that can be dropped into emails, plus the components pinned as company defaults for a slot (currently `footer`). Pass defaultsOnly to see only the pinned defaults. | Solo lectura |
| get_email_component | Get one email component's blocks and metadata by id, including its `version` and whether it is pinned as a company default slot. | Solo lectura |
| get_default_email_component | Get the company's default component for a slot - use this to read the current email footer before editing it. Returns not_found when no default is set, which means emails fall back to the footer Sequenzy generates from the company profile. | Solo lectura |
| preview_default_email_component | Preview a proposed default footer without saving or sending. Provide the proposed blocks and explicit application scopes. Returns affected counts, every affected/skipped email and a token for set_default_email_component. The application scopes select which categories to update: sequences, draft/scheduled campaigns, transactional templates, reusable templates, or future emails only. Customized footers are preserved unless explicitly included. Set renderPreview to get footerHtml and before/after HTML samples, one per selected category; open or save that HTML and use a browser to inspect the visual result. Use sample with an affected id/kind to render a particular email version. Previews are layout-only, without recipient personalization. A stale apply returns conflict; preview again instead of silently retrying with a new token. | Solo lectura |
| set_default_email_component | Create or replace the company's default component for a slot. This is the supported way to change the footer on every email: sequence, campaign, and AI-generated emails built from blocks clone this component when they are built, instead of the generated default footer. Raw HTML emails are the exception - block chrome cannot be injected into raw HTML, so they keep their own markup and a generated unsubscribe footer is appended at send time unless the HTML contains {{unsubscribeUrl}}. Call get_default_email_component first and edit the blocks it returns, otherwise you replace the existing footer wholesale. A default footer always keeps its unsubscribe link enabled - transactional sends hide it at render time. Omitting application keeps existing emails unchanged. To apply to existing emails, call preview_default_email_component first, then pass the same application, blocks and metadata with its token as previewToken. This atomically saves the default and updates only the previewed eligible content. It does not send emails. Customized footers are preserved by default. Conflict means the preview is stale and must be reviewed again. | Acción |
| create_email_component | Create a reusable email component from a block list. Component names are unique per company. Use set_default_email_component instead when the goal is to change the footer every email gets. | Acción |
Configuración
SEQUENZY_API_KEYObligatorioSecretoYour Sequenzy API key (get one from the Sequenzy dashboard under Settings > API Keys)
SEQUENZY_APP_URLOptional Sequenzy app base URL for dashboard links. Defaults to https://sequenzy.com.
| Nombre | Descripción | Tipo |
|---|---|---|
| SEQUENZY_API_KEY | Your Sequenzy API key (get one from the Sequenzy dashboard under Settings > API Keys) | ObligatorioSecreto |
| SEQUENZY_APP_URL | Optional Sequenzy app base URL for dashboard links. Defaults to https://sequenzy.com. |
Verificaciones automáticas
Ejecutado 16 nov 2026Informe completo →Aún no verificado. Un servidor necesita una puntuación de 80, ningún hallazgo grave y suficientes datos para que podamos revisarlo. Cómo funcionan las verificaciones · Qué corregir para obtener más puntos
- Repositorio de origen vinculado y accesiblesequenzy/mcp is public and reachablegithub.com/sequenzy/mcp ↗Aprobado
- El paquete indica este repositorio como su origen@sequenzy/mcp names sequenzy/mcp as sourcenpmjs.com/package/@sequenzy/mcp ↗Aprobado
- Procedencia de la compilación (atestación de Sigstore)No build attestation published for this releaseNo verificado
- Licencia de código abierto (OSI)MITspdx.org/licenses/MIT.html ↗Aprobado
- Commit o lanzamiento en el último añoLast commit or release 2026-11-13github.com/sequenzy/mcp ↗Aprobado
- OpenSSF ScorecardNo verificado
- El nombre del paquete no es similar a un paquete popularNo popular package one edit away from @sequenzy/mcpAprobado
- No hay avisos críticos en el paquete ni en sus dependenciasthe package checked; no critical advisoryosv.dev ↗Aprobado
- Sin instrucciones ocultas para el modeloA tool description points at credential filesHallazgo
- Las herramientas destructivas están etiquetadasDestructive tools carry destructiveHintAprobado
- El endpoint remoto responde al handshake de MCPHTTPS endpoint answered 401 (sign-in required)Aprobado
README
Ver en GitHubSequenzy MCP Server
Official MCP server for Sequenzy, the AI-powered email marketing platform.
Connect Sequenzy to Claude Desktop, Claude Code, Codex, Cursor, Windsurf, VS Code Copilot, OpenClaw, and other MCP clients so your AI assistant can manage email operations with structured tools instead of hand-written API calls.
Interactive email and sequence previews
In clients that support MCP Apps, including compatible ChatGPT and Claude connections, you can review your saved emails and sequences directly in the conversation.
render_emailopens the rendered email with Desktop (640px) and Mobile (375px) views, its subject, preview text and personalization warnings. Pass a campaign ID, template ID, or sequence ID plus node ID using the existing tool inputs. It renders your saved content through Sequenzy's email renderer. Personalization and locale options remain available through the same tool.preview_sequence({ companyId, sequenceId })opens the saved sequence as a connected diagram with triggers, delays, branches and actions. Select an email to render it, switch between available A/B variants, or use the Emails tab to browse the sequence's emails. Random splits show their configured allocation. Refresh reloads saved content, and failed requests offer an explicit retry.list_sequence_email_variations,generate_sequence_email_variations,list_campaign_variationsandgenerate_campaign_variationsopen the design variations side by side, each rendered as it would send. While some are still generating, the view updates every few seconds for up to five minutes; after a generate call it follows only the new ones. Use for Email 1 and Use for every email (campaigns: Use this design) ask for confirmation, then save throughuse_sequence_email_variationoruse_campaign_variationwith your existing permissions and tell your assistant what changed. Nothing changes until you confirm.
In Claude, previews can also open in the side panel or full screen with the expand button in the view's header, and open_sequenzy is listed on Claude's Artifacts page so you can open (and pin) your workspace without asking. These parts of Claude are rolling out gradually; where they are unavailable the previews work as before.
open_sequenzy({ companyId? })opens your workspace: the company's sequences (up to 50) and recent campaigns (up to 25), with a company switcher when your connection reaches more than one. Select a sequence to open its flow or a campaign to render its email; the back link returns to the list. Every input is optional. A list your key cannot read shows as unavailable while the other still loads.- While a preview is open, your assistant can change what it shows through tools the view itself offers:
get_view_state(what you have open),open_sequence_step,open_sequence,open_campaign_email,show_sequence_overview,set_preview_widthandreload_view. After it edits the sequence or email you are looking at,reload_viewshows the saved change in place. These tools only change the view; edits go through the usual write tools and permissions. Your client may ask you to approve them. - Ask for changes, on an open email, sends your assistant a message naming that email so it can suggest improvements. Nothing changes until you agree to an edit.
Supporting workflows use compact cards in the same Sequenzy style:
open_image_upload({ companyId })opens an image picker. Choose a PNG, JPEG, GIF or WebP up to 10 MB, optionally describe it, then select Upload image. Opening the widget does not write anything. Submission usesupload_image_assetwith your existing permissions and returns the saved asset to your assistant. After an uncertain failure, check your image library before retrying; uploads are never automatically retried.monitor_subscriber_import({ companyId, importId })shows counts and progress for an existing import. It pollsget_subscriber_importevery five seconds while running, for up to five minutes. Completion, blocked status, errors or leaving the view stop polling. Refresh status restarts monitoring. This does not start or retry imports.
Both widgets pin the company explicitly. Clients without MCP Apps can use upload_image_asset and get_subscriber_import directly; the same operations are available through the existing REST API and CLI.
The sequence preview requires an explicit company ID from get_account and keeps every email render pinned to that company. Previews never activate sequences, enroll subscribers or send email. A/B previews require the existing access to variant content; unavailable variants are shown as unavailable instead of being replaced with the control email.
Email content is isolated from the widget. Scripts, forms, nested frames and link navigation are disabled; external stylesheets and fonts are not loaded. HTTPS email images may load from their original hosts, subject to the client's policy. The widget itself bundles its logo and font and uses authenticated MCP tool calls without exposing credentials. Actual email-client rendering can differ, so use test sends for final inbox verification.
Clients without MCP Apps still receive the existing rendered HTML or sequence data. The same workflows are available through the existing public render and sequence-read endpoints and CLI render/sequence commands. No new REST endpoint or CLI command is needed for this presentation layer.
What You Can Do
Use create_campaign_for_audience for a frozen group of selected contacts or all contacts matching contact/activity filters. It creates a blank draft, reports selected and currently eligible counts, and preserves existing campaign read/update/clear/schedule workflows. The limit is 100,000 contacts and 8 MiB; drafting sends no email.
- Manage subscribers, tags, lists, and dynamic segments, including bulk tag reconciliation and synthetic event testing.
- Sync segments to Meta custom audiences for Facebook and Instagram retargeting.
- Sync contacts and events from Snowflake, BigQuery, Redshift or Postgres on a schedule.
- Stream email events, custom events and subscriber snapshots to your S3 or GCS bucket for your data warehouse.
- Manage products and attach digital delivery files for purchase automations.
- Upload hosted email images with alt text and reusable responsive crop settings.
- Draft, update, schedule, and inspect campaigns, including resolved audience previews, persisted conversion goals, and From, Reply-To, CC, and BCC identities.
- Render campaigns, sequence steps, and templates to their exact email-safe HTML without sending.
- Add one-click Poll and NPS survey blocks to emails and inspect campaign response summaries.
- Create and edit email sequences, including multi-list/tag triggers, entry-audience and property-filtered stop conditions, sending identity overrides, existing graph restructuring, and direct step test sends to internal reviewers.
- Cancel, pause, resume, duplicate, or delete campaigns and enroll contacts into sequences.
- Manage transactional email templates and send transactional emails to shared To, Cc, and Bcc recipient lists.
- Supply localized template variants or queue AI translation for enabled locales.
- Create, preview, edit, publish, unpublish, and delete landing pages.
- Create list-scoped saved signup forms with responsive stack, row, grid, and single-image overlay block groups (including foreground gap controls), then return client-safe static-site embeds.
- Create, target, publish, duplicate, and deploy saved signup popups with the same recursive block layouts.
- Connect and verify custom domains for published landing pages.
- Manage team invitations, the inbox (its address, conversations, replies, forwards), and outbound webhook endpoints.
- Manage geofences (places the mobile SDKs watch) and bulk-import them by external ID. Visits become
app.geofence_*events for sequence triggers andgeofencesegment filters. - Generate email copy, subject lines, and multi-step sequences.
- Inspect analytics, subscriber activity, deliverability health, company-level sending pauses, integrations, published event payload schemas, sending identities, tracking settings, and dashboard URLs.
- Inspect whether "Sent with Sequenzy" is visible for a workspace, why the owner subscription does or does not remove it, and open the canonical subscription page for an upgrade or renewal. Entitlement changes apply to future sends from existing live sequences without editing their blocks.
- Diagnose why sending is paused and restore eligible hard-bounce pauses after confirming list cleanup.
- Inspect exact-recipient bounce, complaint, and email-hygiene suppression, and clean up eligible stale bounces without exposing the shared SES suppression list.
- Configure company product info, account-wide sending identity defaults, rename individual sender and reply-to profiles, manage sender domains, and inspect integration examples for common frameworks.
Every published MCP tool includes explicit readOnlyHint, destructiveHint, idempotentHint, and openWorldHint annotations so compatible clients can display accurate tool-use affordances. Any write that overwrites, cancels, deletes, changes a live sequence, or can send email or SMS is marked destructive; tools that can email real contacts, publish a public link, or reach a third-party service are marked open-world. Tools also publish outputSchema definitions (except on the OpenAI-reviewed route) and return structuredContent, giving clients and models machine-readable result shapes for follow-up calls.
Tool results and recovery
Tool failures retain isError: true and readable error text, and also return
structuredContent with { success: false, error: { code, message, howToFix, docsUrl } }. HTTP errors include statusCode; a valid API Retry-After header
adds retryAfterSeconds. These fields are omitted when unavailable. The OpenAI
profile uses safe recovery descriptions and known codes or status-derived
fallbacks, omitting raw backend diagnostics.
Use the error code and recovery guidance to choose your next step. The client
retries a read (GET) up to 3 times within about 3 seconds after a 502, 503, 504
or network error, waiting out a Retry-After of 5 seconds or less, and never
retries a write. After an uncertain send_email outcome, reuse the
original idempotencyKey with unchanged input and inspect emailSendId when
available. A retry hint alone does not mean repeating a write is safe.
Segment counts and segment-filtered reads fail with
SEGMENT_REFERENCE_DEPTH_EXCEEDED or SEGMENT_REFERENCE_COUNT_EXCEEDED when
segments reference other segments too deeply or too many times. Retrying does
not help; reduce the nesting or the number of references with
update_segment first.
Successful results keep their existing shapes. Recoverable partial subscriber
event imports retain their counts and row failures without isError; inspect
those rows before retrying failed records.
Subscriber writes (add_subscriber, update_subscriber with tag changes,
trigger_subscriber_event, trigger_subscriber_events,
import_subscriber_events) report sequenceTriggersRan. When it is false
with sequenceTriggersSkippedReason: "missing_scope", the key lacks
automations:trigger: the write was saved but no sequence started, and
warnings says so.
When an account key runs a tool with no selected company and no companyId,
the API uses a default company. The result then gets a second text block that
names that company; structuredContent is unchanged. Call select_company
when the key reaches more than one company.
Quick Setup
The easiest setup path is the Sequenzy wizard:
npx @sequenzy/setup
The wizard opens the browser login flow, creates a personal API key, detects supported AI clients, and configures them automatically when possible.
Hosted Remote MCP
For clients that support Streamable HTTP MCP, use Sequenzy's hosted endpoint instead of running a local stdio process:
https://api.sequenzy.com/v1/mcp
The Sequenzy plugin in the ChatGPT plugin directory uses the reviewed hosted surface, which ChatGPT custom connectors can also add directly:
https://api.sequenzy.com/v1/mcp/openai
That surface shares the same implementation and keeps the standard tool set
except for operations that accept or manage credentials, or return unprojected
payloads: connect_integration, create_api_key, list_api_keys,
update_api_key, revoke_api_key, delete_api_key,
request_api_key_handoff, create_data_export, update_data_export,
create_warehouse_connection, update_warehouse_connection,
preview_warehouse_query, create_webhook, the geofence tools
(list_geofences, get_geofence, create_geofence, update_geofence,
delete_geofence, import_geofences; they carry place coordinates), list_webhook_deliveries,
list_capture_submissions, replay_webhook_delivery, and
rotate_sequence_inbound_webhook_secret. It also omits add_website, a
compatibility alias of add_sending_domain, and submit_feedback. API keys, including the one behind
the ChatGPT connection, are managed in the dashboard. The surface also omits
subscription and upgrade links: subscriptionUrl fields, the
get_app_urls subscription URL, and the emailBranding tier and upgrade
action are removed, and get_app_urls rejects the billing and
subscription settings tab aliases. The surface also omits the push credential
tools set_apns_credentials and set_fcm_credentials, which take customer
private keys as input, and register_push_device, an app-backend call whose
required device token the surface's credential policy rejects.
Four tools use clearer names on this surface; standard MCP and the stdio package keep the original names:
| Standard MCP | OpenAI-reviewed route |
|---|---|
list_websites |
list_sending_domains |
check_website |
get_sending_domain |
resend_campaign_to_non_openers |
create_non_opener_resend_draft |
update_sms_number_label |
update_sms_number_settings |
edit_sequence_graph is listed there as four single-operation tools with the
same validation and behavior: move_sequence_step (move_node),
duplicate_sequence_step (duplicate_node), delete_sequence_step
(delete_node) and replace_sequence_paths (replace_edges).
OpenAI reviews again any published tool whose definition changes, so two
filters keep their approved definitions there until the next plugin
submission: list_email_sends describes bounceType as Permanent or
Transient (Soft and Blocked are still accepted), and
list_sequence_enrollments offers the active, waiting, completed, failed and
cancelled statuses and defaults to active and waiting, so paused enrollments
are not listed there.
The preview widget on this surface lists the image origins it may load
(https://images.sequenzy.com and the Pexels and Unsplash image hosts), so
other external images in an email preview show their alt text instead.
Tool annotations disclose supported replacement and irreversible-delivery modes, including automation-triggering actions. Edits to email content that can be live or shown on a public share link are destructive and open-world. Public content, provider changes, tools that fetch URLs or call outside services (link checks, AI drafting that can read URLs, gallery ranking), and membership changes that can start segment-triggered sequences are marked open-world; private draft creation and additive notes remain non-destructive. On the ChatGPT surface each tool's annotation justifications live in src/tools/openai-justifications/, and a tool whose description does not already explain a hint gets an effects sentence appended. Feedback goes to the Sequenzy team through Telegram with the authenticated company name/ID and account email added for attribution, which clients must disclose before sending.
To keep its tool list small, this surface publishes no output schemas (results
still include structuredContent).
On every surface, sequence tools describe repeated nested step schemas once and
point later fields at the first copy ("Same schema as branches[].steps in this
tool."). Those fields accept the same input as before; the server validates
nested steps.
Meta's Muse connector uses a smaller hosted surface, because Muse rejects large tool lists:
https://api.sequenzy.com/v1/mcp/muse
It serves about 70 everyday tools from the standard surface (account,
subscribers, lists, tags, segments, campaigns, sequences, templates,
transactional sends, stats and the inbox) with the same names, inputs and
handlers, and keeps tools/list under 200 KB by leaving out output schemas.
Tool results still include structuredContent. The large sequence builders
(create_sequence, update_sequence, insert_sequence_step) are left out;
generate_sequence, create_sequence_from_example and edit_sequence_emails
cover the same work. Calls to tools outside the list are rejected. The list is
MUSE_TOOL_NAMES in src/tools/muse-profile.ts.
Remote clients should authenticate with the Sequenzy OAuth flow when supported. Local and automation clients can still use the stdio package below with SEQUENZY_API_KEY.
The hosted endpoint and the stdio package support MCP specification
2026-07-28 while remaining compatible with 2025-era clients. Modern HTTP
clients use per-request discovery and method headers; existing clients keep
working through the same endpoint and package command.
Machine-readable discovery files:
- MCP server manifest:
server.json - Agent card:
.well-known/agent-card.json - Agent capability manifest:
agent-capability.json - OpenClaw skill metadata:
openclaw/skill.json
Preguntas sobre Sequenzy MCP
¿Cómo agrego Sequenzy MCP a Claude Code?
Ejecuta claude mcp add --env 'SEQUENZY_API_KEY=<YOUR_SEQUENZY_API_KEY>' mcp -- npx -y @sequenzy/mcp en tu terminal. Añade --scope user para que esté disponible en todos los proyectos.
¿Cómo agrego Sequenzy MCP a Cursor?
Abre ~/.cursor/mcp.json y añade el servidor en mcpServers, usando la configuración de Cursor que se muestra en la sección de instalación de arriba.
¿Qué herramientas proporciona Sequenzy MCP?
Expone 226 herramientas, incluidas open_sequenzy, preview_sequence, open_image_upload, monitor_subscriber_import, get_account.
¿Puede el MCP server enviar correos electrónicos reales?
Las herramientas de vista previa específicamente no activan secuencias, no inscriben suscriptores ni envían correos; sirven para renderizar y visualizar contenido.
¿Qué sucede si falla una llamada a una herramienta?
Los fallos de las herramientas devuelven un estado de error con un mensaje y orientación sobre cómo solucionarlo, incluyendo una URL de documentación y, potencialmente, una duración de reintento.
¿Puedo previsualizar cómo se ve un correo en dispositivos móviles?
Sí, la herramienta render_email proporciona vistas tanto de escritorio (640px) como móviles (375px).