Permet aux acheteurs média de gérer les campagnes, les zones, les membres de l'équipe et les statistiques de trafic directement via une interface de chat.
Aussi connu sous affset/mcp, mcp, @affset/mcp
- Étoiles
- 0
- Outils
- 31
- Téléchargements, 30 jours
- 258
- Dernière version
- 0.4.0
- Dernier commit
- 13 sept. 2026
- Licence
- MIT
Qu'est-ce que affset MCP ?
affset est un serveur MCP. Permet aux acheteurs média de gérer les campagnes, les zones, les membres de l'équipe et les statistiques de trafic directement via une interface de chat. Il expose 31 outils, comme whoami, get_stats, list_campaigns.
Résumé généré à partir du README du projet
À propos du MCP server affset
Les acheteurs média peuvent gérer l'intégralité de leur flux de travail publicitaire via une interface de chat grâce à ce MCP server. Il remplace la navigation traditionnelle dans les tableaux de bord en permettant aux utilisateurs d'extraire des statistiques de trafic, de gérer des campagnes et de contrôler les accès de l'équipe en utilisant le langage naturel. Le serveur se connecte à l'API du tenant affset, offrant un moyen de surveiller le ROI, les conversions et les coûts média selon diverses dimensions comme le pays, l'annonceur ou l'éditeur.
Les utilisateurs peuvent choisir entre une connexion distante hébergée via OAuth ou une installation locale utilisant un package npm avec un Bearer token et un namespace. L'endpoint hébergé est compatible avec Claude, Cursor et Claude Code. Pour les configurations locales, le serveur fonctionne via stdio et peut être configuré pour un accès en lecture seule. L'ensemble d'outils comprend des commandes spécifiques pour gérer les sources de trafic, ajuster les enchères réseau pour des plateformes comme ExoClick ou TrafficStars, et créer de nouvelles zones ou de nouveaux membres d'équipe avec des protections de type dry-run intégrées pour éviter les modifications accidentelles.
Ce que vous pouvez faire avec
- 1Analyser la performance des campagnesDemandez un tableau de statistiques de trafic incluant les clics, les conversions et le ROI regroupés par dimensions spécifiques.
- 2Modifier les enchères réseauMettez à jour l'enchère en USD pour une campagne spécifique sur une source de trafic comme RichAds ou ExoClick.
- 3Gérer les accès de l'équipeInvitez un nouveau membre de l'équipe sur le compte et attribuez-lui un rôle spécifique.
- 4Automatiser la configuration des campagnesCréez une nouvelle campagne avec une URL d'offre spécifique, un ciblage géographique et un taux de payout.
Extrait du README du projet. Chaque cas d'utilisation est vérifié par rapport à une citation du document.
Comment installer affset MCP
Formats vérifiés le 10 nov. 2026Ajouter à CursorAjouter à GooseInstaller affset MCP dans Claude Code
$ claude mcp add --env 'AFFSET_API_KEY=<YOUR_AFFSET_API_KEY>' --env 'AFFSET_BASE_URL=<AFFSET_BASE_URL>' --env 'AFFSET_NAMESPACE=<AFFSET_NAMESPACE>' mcp -- npx -y @affset/mcpInstaller affset MCP dans Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@affset/mcp"
],
"env": {
"AFFSET_API_KEY": "<YOUR_AFFSET_API_KEY>",
"AFFSET_BASE_URL": "<AFFSET_BASE_URL>",
"AFFSET_NAMESPACE": "<AFFSET_NAMESPACE>"
}
}
}
}Installer affset MCP dans VS Code
.vscode/mcp.json
{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@affset/mcp"
],
"env": {
"AFFSET_API_KEY": "<YOUR_AFFSET_API_KEY>",
"AFFSET_BASE_URL": "<AFFSET_BASE_URL>",
"AFFSET_NAMESPACE": "<AFFSET_NAMESPACE>"
}
}
}
}Installer affset MCP dans VS Code CLI
$ code --add-mcp '{"name":"mcp","type":"stdio","command":"npx","args":["-y","@affset/mcp"],"env":{"AFFSET_API_KEY":"<YOUR_AFFSET_API_KEY>","AFFSET_BASE_URL":"<AFFSET_BASE_URL>","AFFSET_NAMESPACE":"<AFFSET_NAMESPACE>"}}'Installer affset MCP dans Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@affset/mcp"
],
"env": {
"AFFSET_API_KEY": "<YOUR_AFFSET_API_KEY>",
"AFFSET_BASE_URL": "<AFFSET_BASE_URL>",
"AFFSET_NAMESPACE": "<AFFSET_NAMESPACE>"
}
}
}
}Installer affset MCP dans Codex
$ codex mcp add mcp --env 'AFFSET_API_KEY=<YOUR_AFFSET_API_KEY>' --env 'AFFSET_BASE_URL=<AFFSET_BASE_URL>' --env 'AFFSET_NAMESPACE=<AFFSET_NAMESPACE>' -- npx -y @affset/mcpInstaller affset MCP dans Codex config.toml
~/.codex/config.toml
[mcp_servers.mcp] command = "npx" args = ["-y", "@affset/mcp"] [mcp_servers.mcp.env] AFFSET_API_KEY = "<YOUR_AFFSET_API_KEY>" AFFSET_BASE_URL = "<AFFSET_BASE_URL>" AFFSET_NAMESPACE = "<AFFSET_NAMESPACE>"
Installer affset MCP dans Gemini CLI
$ gemini mcp add -s user -e 'AFFSET_API_KEY=<YOUR_AFFSET_API_KEY>' -e 'AFFSET_BASE_URL=<AFFSET_BASE_URL>' -e 'AFFSET_NAMESPACE=<AFFSET_NAMESPACE>' mcp npx -- -y @affset/mcpInstaller affset MCP dans Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@affset/mcp"
],
"env": {
"AFFSET_API_KEY": "<YOUR_AFFSET_API_KEY>",
"AFFSET_BASE_URL": "<AFFSET_BASE_URL>",
"AFFSET_NAMESPACE": "<AFFSET_NAMESPACE>"
}
}
}
}Outils MCP affset 31
Détectés depuis le READMEwhoamiget_statslist_campaignsget_campaignlist_zoneslist_teamcreate_team_membercreate_campaignset_campaign_statusupdate_campaigncreate_zoneupdate_zone
| whoami | — | |
| get_stats | — | |
| list_campaigns | — | |
| get_campaign | — | |
| list_zones | — | |
| list_team | — | |
| create_team_member | — | |
| create_campaign | — | |
| set_campaign_status | — | |
| update_campaign | — | |
| create_zone | — | |
| update_zone | — |
Afficher les 31 outilsAfficher moins d'outils
list_traffic_sourcescreate_traffic_sourceupdate_traffic_sourcelist_source_bidsset_source_bidget_zone_urlget_tracking_linkcut_zoneslist_payout_rulesset_payout_ruledelete_payout_ruleset_payout_goallist_targeting_typeslist_targeting_rulesset_targeting_ruleremove_targeting_rulelist_sub_labelsset_sub_labelslist_conversions
| list_traffic_sources | — | |
| create_traffic_source | — | |
| update_traffic_source | — | |
| list_source_bids | — | |
| set_source_bid | — | |
| get_zone_url | — | |
| get_tracking_link | — | |
| cut_zones | — | |
| list_payout_rules | — | |
| set_payout_rule | — | |
| delete_payout_rule | — | |
| set_payout_goal | — | |
| list_targeting_types | — | |
| list_targeting_rules | — | |
| set_targeting_rule | — | |
| remove_targeting_rule | — | |
| list_sub_labels | — | |
| set_sub_labels | — | |
| list_conversions | — |
Configuration
AFFSET_API_KEYRequisSecretTenant API key (Bearer token). Its namespace must match AFFSET_NAMESPACE.
AFFSET_BASE_URLRequisOrigin of your affset ad server API (no path, query, or embedded credentials).
AFFSET_NAMESPACERequisTenant namespace this key belongs to (lowercase letters, numbers, hyphens; 3–63 chars).
AFFSET_READ_ONLYOptional. Set to true to register only read-only tools (no create/update/delete/cut).
AFFSET_REQUEST_TIMEOUT_MSOptional per-request HTTP timeout in milliseconds (1000–300000).
| Nom | Description | Type |
|---|---|---|
| AFFSET_API_KEY | Tenant API key (Bearer token). Its namespace must match AFFSET_NAMESPACE. | RequisSecret |
| AFFSET_BASE_URL | Origin of your affset ad server API (no path, query, or embedded credentials). | Requis |
| AFFSET_NAMESPACE | Tenant namespace this key belongs to (lowercase letters, numbers, hyphens; 3–63 chars). | Requis |
| AFFSET_READ_ONLY | Optional. Set to true to register only read-only tools (no create/update/delete/cut). | |
| AFFSET_REQUEST_TIMEOUT_MS | Optional per-request HTTP timeout in milliseconds (1000–300000). |
Vérifications automatisées
Exécuté le 12 nov. 2026Rapport complet →A passé nos vérifications automatisées. Il ne s'agit pas d'un audit de sécurité : lisez le code et donnez au serveur l'accès le plus restreint dont il a besoin. Fonctionnement des vérifications
- Dépôt source lié et accessibleRéussi
- Licence open source (OSI)Réussi
- Commit ou version publiée au cours de l'année écouléeRéussi
- Nom du package non proche d'un package populaireRéussi
- L'endpoint distant utilise HTTPSRéussi
- Les outils destructifs sont étiquetésNon vérifié
- Aucune instruction cachée destinée au modèleRéussi
- Aucun avis critique dans le package ou ses dépendancesRéussi
README
Voir sur GitHubaffset MCP server
An MCP server that lets a media buyer run affset from a chat client — pull stats, manage campaigns/zones/team, payouts, targeting, sub labels, and cut underperforming zones in plain language, no dashboard.
The tools wrap the existing affset tenant API. Connect through the hosted
endpoint (OAuth, no API key) or run this package locally (Bearer token +
X-Namespace). One connection serves one tenant.
Fastest way to connect — the hosted endpoint. Add
https://mcp.affset.com/mcp as a remote MCP server in Claude (web or desktop),
Cursor, Claude Code, or any client that supports streamable HTTP with OAuth:
paste the URL, sign in with your affset email, keep Read-only (the consent
default) or grant full access. Every connection shows up on the dashboard
Integrations page and can be revoked individually. Setup guide:
affset.com/integrations.
The npm package below is the self-host path: same tool roster, runs on your
machine with an API key you manage yourself. Stdio defaults to full access
unless you set AFFSET_READ_ONLY=true.
Tools
| Tool | What it does |
|---|---|
whoami |
Show the tenant this server is bound to: namespace, API base, derived dashboard URL, and (when readable) company / timezone / custom API domain. Read-only. |
get_stats |
Traffic stats grouped by a dimension (date, campaign, zone, country, sub1–5, advertiser, publisher, …), optionally narrowed with advertiser_email/publisher_email filters (one user, any group_by). Returns clicks, conversions, CR, payout, media cost and ROI as a table. paid_only defaults to true (same as the dashboard) so CR excludes informative conversions. Sub columns use the tenant's sub labels when configured. Per-user groupings and filters are owner/manager plus the matching side's manager role. |
list_campaigns |
List campaigns (status / name filter, pagination). |
get_campaign |
One campaign's full record — every field (untruncated offer URL, exact schedule, budgets/pacing, silent flag, payout goal type) plus its targeting rules and payout rules, in one call. |
list_zones |
List traffic-source zones (status / name filter, pagination, linked source). |
list_team |
List team members (email, role, manager). Never returns API tokens. |
create_team_member |
Invite a team member (owner, manager, publisher, advertiser, publisher_manager, advertiser_manager). A scoped manager key can only create its own managed role, self-assigned. Returns the new API key once — list_team never shows it again. Dry-run by default; confirm: true to apply. |
create_campaign |
Create a campaign from an advertiser email, offer URL, geo, payout and name. Defaults: CPA / rate 0, paused, global payout rule, and a ready tracking link (linked source template when configured; otherwise source_click_id={clickid} + sub placeholders). Dry-run by default; confirm: true to apply. |
set_campaign_status |
Run or pause a campaign (action: "run" | "pause"). Dry-run by default; confirm: true to apply. Running can hit the plan's active-campaign limit. |
update_campaign |
Partial update (name, offer URL, status, rate, budgets, dates, …). Dry-run by default; confirm: true to apply. Prefer set_campaign_status for run/pause. |
create_zone |
Create a traffic-source zone (name + optional postback/site/traffic-back URLs, optional traffic_source_id link). Always created active. Dry-run by default; confirm: true to apply. |
update_zone |
Partial update (name, status, URLs, traffic_source_id). Dry-run by default; confirm: true to apply. Pass null to clear a URL or unlink the source. |
list_traffic_sources |
List traffic sources — the networks bought from, each with the tracking/postback templates its linked zones use. API token shown only as set/none. |
create_traffic_source |
Create a traffic source, optionally from a network preset (exoclick, trafficstars, propellerads, adsterra, richads) that copies verified templates into an editable row. Dry-run by default; confirm: true to apply. |
update_traffic_source |
Partial update (name, templates, api_token, status). Linked zones pick the new tracking template up immediately. Dry-run by default; confirm: true to apply. |
list_source_bids |
The network campaigns of one traffic source with their current bids, read live from the network account (ExoClick, TrafficStars, RichAds — needs the source's API token): status, pricing model, bid in USD, last change that entered Affset's mutation stage. Read-only. |
set_source_bid |
Set one network campaign's bid in USD using its existing pricing model (for RichAds: CPM for pops, CPC for push/display). The dry run returns expected_current_bid; pass that value back with confirm: true to bind the write to the bid reviewed, so a concurrent change is refused rather than overwritten. Records the attempt in the source's bid history and reports applied only when the network echoes the new value. Raising a bid above 5× needs allow_large_increase: true. Dry-run by default. |
get_zone_url |
The /serve URL to paste into a network's campaign settings — rotates across the zone's active campaigns. A zone linked to a traffic source renders that source's tracking template; otherwise prefilled sub convention + optional cost macro. Warns when no active campaigns are visible. |
get_tracking_link |
The /track/click link for an existing campaign + zone — straight to one active campaign, with no rotation or targeting checks. Renders a linked source's template like get_zone_url. Re-derives what create_campaign echoed on create. |
cut_zones |
Blacklist underperforming zones on a campaign by threshold (CR / spend / ROI). Dry-run by default; confirm: true to apply. |
list_payout_rules |
List a campaign's global + per-zone payout rules and its payout_goal_type. |
set_payout_rule |
Upsert a global or zone-specific payout. Dry-run by default; confirm: true to apply. |
delete_payout_rule |
Delete a global or zone-specific payout rule. Dry-run by default; confirm: true to apply. |
set_payout_goal |
Set or clear payout_goal_type (goal-based conversions). Dry-run by default; confirm: true to apply. |
list_targeting_types |
Catalog of targeting rule types, flagging the seeded ones /serve never evaluates. |
list_targeting_rules |
List a campaign's targeting rules, flagging any that have no effect. |
set_targeting_rule |
Upsert one targeting rule (safe merge), normalised to what /serve matches. Dry-run by default; confirm: true to apply. |
remove_targeting_rule |
Remove one targeting rule by id or type+method. Dry-run by default; confirm: true to apply. |
list_sub_labels |
List tenant display names for sub1–sub5. |
set_sub_labels |
Set or clear sub labels (partial; null clears). Dry-run by default; confirm: true to apply. |
list_conversions |
List conversion audit records (payout, spend, pixel type, payload, postback). paid_only filters server-side; other optional filters are client-side on the current page. |
Which URL do I give the network?
get_zone_url (/serve/{zone}) |
get_tracking_link (/track/click/{campaign}/{zone}) |
|
|---|---|---|
| Picks the campaign | affset, from the zone's rotation | you, one fixed campaign |
| Needs an active campaign | yes — otherwise traffic back / unsold | yes — otherwise 404 |
| Needs an active zone | yes | yes |
| Geo & targeting rules | enforced | not enforced |
cost= lands on |
the impression row | the click row |
Use one or the other for a given traffic stream — never both with cost=, or the
media cost is counted twice.
Both use the tenant's custom API domain when one is set, since the URL gets pasted
into the network verbatim. Macros ({clickid}, [CLICK_ID], ${SUBID}) are inserted
without percent-encoding — the source expands them before the request reaches affset.
cut_zones only ever adds zones to a campaign's blacklist, and does a
read-merge-write so existing targeting rules are never touched.
create_campaign needs a traffic-source zone for the tracking link: pass
zone_id, or let it auto-pick when the namespace has exactly one active zone.
Campaigns are created paused; activate them before sending traffic through either
URL. Both URL types also require an active zone. Geo whitelist is enforced in /serve
only — the direct tracking link is not geo-gated, but it still requires an active,
currently serviceable campaign.
Documentation resources
Beyond the tools, the server exposes the affset API reference as MCP
resources, so an
assistant can answer "how does conversion tracking work?" or "what does /serve
accept?" from the docs themselves — not just from the tool schemas.
| Resource URI | Type | Content |
|---|---|---|
affset://docs/api-reference |
text/markdown |
The full API reference — endpoints, auth, roles, examples. |
affset://docs/api-reference.json |
application/json |
The same reference as structured data, for programmatic use. |
They're the exact content published at affset.com/docs,
generated from one source, and fetched at read time from AFFSET_DOCS_URL
({origin}/api-reference.md and {origin}/api-reference.json) — so they always
reflect the currently published docs, not a copy pinned to this package. The
fetch sends no credentials (the docs are public and live on a different
origin than the tenant API). HTML SPA fallbacks, redirects, invalid JSON, and
oversized bodies are rejected. Both resources are always available, including
under AFFSET_READ_ONLY.
Questions sur affset MCP
Comment ajouter affset MCP à Claude Code ?
Exécutez claude mcp add --env 'AFFSET_API_KEY=<YOUR_AFFSET_API_KEY>' --env 'AFFSET_BASE_URL=<AFFSET_BASE_URL>' --env 'AFFSET_NAMESPACE=<AFFSET_NAMESPACE>' mcp -- npx -y @affset/mcp dans votre terminal. Ajoutez --scope user pour le rendre disponible dans tous les projets.
Comment ajouter affset MCP à Cursor ?
Ouvrez ~/.cursor/mcp.json et ajoutez le serveur sous mcpServers, avec la configuration Cursor indiquée dans la section d'installation ci-dessus.
Quels outils affset MCP fournit-il ?
Il expose 31 outils, dont whoami, get_stats, list_campaigns, get_campaign, list_zones.
Puis-je exécuter ce serveur localement ?
Oui, vous pouvez exécuter le package npm localement en utilisant un Bearer token et un X-Namespace.
Quel type de statistiques puis-je récupérer ?
Vous pouvez extraire des statistiques de trafic incluant les clics, les conversions, le CR, le payout, le coût média et le ROI, regroupés par dimensions telles que la date, la campagne ou le pays.