MCP server Pingstack
Commit 10 de nov. de 2026Permite que um cliente de IA monitore cron jobs, sites, DNS, certificados SSL e expiração de domínios, além de gerenciar dashboards e alertas.
Também conhecido como atomicvan/pingstack-mcp, pingstack-mcp
- Stars
- 0
- Ferramentas
- 33
- Downloads, 30 dias
- —
- Versão mais recente
- 1.2.0
- Último commit
- 10 de nov. de 2026
- Licença
- MIT
O que é o MCP Pingstack?
Pingstack é um MCP server. Permite que um cliente de IA monitore cron jobs, sites, DNS, certificados SSL e expiração de domínios, além de gerenciar dashboards e alertas. Ele expõe 33 ferramentas, como get_account_capabilities, list_checks, get_check.
Resumo gerado a partir do README do projeto
Sobre o MCP server de Pingstack
O Pingstack oferece um MCP server remoto que permite que agentes de codificação gerenciem tarefas abrangentes de monitoramento. Desenvolvedores podem utilizá-lo para supervisionar cron jobs por meio de verificações de heartbeat, monitorar a disponibilidade de sites, verificar portas TCP e acompanhar registros DNS, certificados SSL e expiração de domínios. O servidor é hospedado no Pingstack, o que significa que não é necessária instalação local, e ele se conecta via streamable HTTP usando OAuth no navegador ou uma API key.
Os usuários podem interagir com 35 ferramentas diferentes para gerenciar incidentes, grupos de notificação e dashboards. O sistema suporta a criação de dashboards privados ou a publicação de páginas de status públicas. Embora o serviço seja gratuito para até 15 verificações, ele se integra com diversos clientes como Claude Code, Cursor e GitHub Copilot CLI. Também oferece Agent Skills especializadas para Claude Code e Cursor para automatizar fluxos de trabalho complexos, como configurar o monitoramento de cron jobs ou gerar relatórios de uptime.
O que você pode fazer com ele
- 1Automatizar o monitoramento de cron jobsUm agente pode ler o agendamento de um job e criar uma verificação de heartbeat correspondente com um período de tolerância específico.
- 2Gerar páginas de status para clientesVocê pode instruir uma IA a construir um dashboard para um cliente específico e publicá-lo como uma página de status pública.
- 3Analisar histórico de incidentesDesenvolvedores podem solicitar detalhes sobre interrupções de serviço recentes, incluindo suas causas e durações.
Escrito a partir do README do projeto. Cada caso de uso é verificado com uma citação dele.
Como instalar o MCP Pingstack
Formatos verificados em 10 nov 2026Adicionar ao CursorAdicionar ao GooseInstalar o MCP Pingstack no Claude Code
$ claude mcp add --transport http pingstack-mcp https://app.pingstack.io/mcpInstalar o MCP Pingstack no Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"pingstack-mcp": {
"url": "https://app.pingstack.io/mcp"
}
}
}Instalar o MCP Pingstack no VS Code
.vscode/mcp.json
{
"servers": {
"pingstack-mcp": {
"type": "http",
"url": "https://app.pingstack.io/mcp"
}
}
}Instalar o MCP Pingstack no VS Code CLI
$ code --add-mcp '{"name":"pingstack-mcp","type":"http","url":"https://app.pingstack.io/mcp"}'Instalar o MCP Pingstack no Codex
$ codex mcp add pingstack-mcp --url https://app.pingstack.io/mcpInstalar o MCP Pingstack no Codex config.toml
~/.codex/config.toml
[mcp_servers.pingstack-mcp] url = "https://app.pingstack.io/mcp"
Instalar o MCP Pingstack no Gemini CLI
$ gemini mcp add -s user --transport http pingstack-mcp https://app.pingstack.io/mcpInstalar o MCP Pingstack no Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"pingstack-mcp": {
"url": "https://app.pingstack.io/mcp",
"type": "http"
}
}
}Ferramentas do MCP Pingstack 33
Detectadas a partir do READMEget_account_capabilitieslist_checksget_checkcreate_checkupdate_checkget_check_historylist_check_runsget_check_metricsget_check_uptimelist_incidentsget_incidentcreate_incident_update
| get_account_capabilities | — | |
| list_checks | — | |
| get_check | — | |
| create_check | — | |
| update_check | — | |
| get_check_history | — | |
| list_check_runs | — | |
| get_check_metrics | — | |
| get_check_uptime | — | |
| list_incidents | — | |
| get_incident | — | |
| create_incident_update | — |
Mostrar todas as 33 ferramentasMostrar menos ferramentas
list_notification_groupslist_account_memberslist_rosterscreate_notification_groupupdate_notification_groupassign_notification_grouplist_dashboardsget_dashboardcreate_dashboardupdate_dashboardattach_check_to_dashboarddetach_check_from_dashboardreorder_dashboard_checkscreate_dashboard_sectionupdate_dashboard_sectionpublish_dashboardget_check_ping_urlsdelete_checkdelete_dashboarddelete_notification_groupunpublish_dashboard
| list_notification_groups | — | |
| list_account_members | — | |
| list_rosters | — | |
| create_notification_group | — | |
| update_notification_group | — | |
| assign_notification_group | — | |
| list_dashboards | — | |
| get_dashboard | — | |
| create_dashboard | — | |
| update_dashboard | — | |
| attach_check_to_dashboard | — | |
| detach_check_from_dashboard | — | |
| reorder_dashboard_checks | — | |
| create_dashboard_section | — | |
| update_dashboard_section | — | |
| publish_dashboard | — | |
| get_check_ping_urls | — | |
| delete_check | — | |
| delete_dashboard | — | |
| delete_notification_group | — | |
| unpublish_dashboard | — |
Verificações automatizadas
Executado em 14 de nov. de 2026Relatório completo →Ainda não verificado. Um servidor precisa de uma pontuação de 80, nenhum problema grave e dados suficientes para verificarmos. Como as verificações funcionam · O que corrigir para ganhar mais pontos
- Repositório de origem vinculado e acessívelatomicvan/pingstack-mcp is public and reachablegithub.com/atomicvan/pingstack-mcp ↗Passou
- O pacote indica este repositório como sua origemNão verificado
- Proveniência da build (atestação Sigstore)Não verificado
- Licença de código aberto (OSI)MITspdx.org/licenses/MIT.html ↗Passou
- Commit ou release no último anoLast commit or release 2026-11-10github.com/atomicvan/pingstack-mcp ↗Passou
- OpenSSF ScorecardNão verificado
- Nome do pacote não é próximo a um pacote popularNão verificado
- Nenhum aviso crítico no pacote ou em suas dependênciasNão verificado
- Sem instruções ocultas para o modelo33 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPassou
- Ferramentas destrutivas estão rotuladasNão verificado
- O endpoint remoto responde ao handshake do MCPHTTPS endpoint answered 401 (sign-in required)Passou
README
Ver no GitHubPingstack MCP server
Pingstack monitors cron jobs and scheduled tasks with heartbeat checks, and checks websites, TCP ports, DNS records, SSL certificates and domain expiry. Its MCP server lets your coding agent set up that monitoring and read the results for you.
Ask your agent to add a heartbeat check for the backup job it just wrote, with the right cron schedule, time zone and grace period. Ask which checks had incidents last week, why, and how long they were down. Ask it to build a dashboard for a client, publish it as a status page and set up who gets alerted.
This repository holds the setup files and directory listings for the hosted server. The server itself runs at Pingstack, so there's nothing to install or run locally.
| Endpoint | https://app.pingstack.io/mcp |
| Transport | Streamable HTTP (stateless, JSON responses) |
| Sign-in | OAuth in your browser, or an API key |
| Price | Free for 15 checks. MCP is included on every plan |
| Docs | docs.pingstack.io/mcp |
Connect
You need a Pingstack account. Sign up free, then add the server to your client. Leave out any Authorization header: the client opens your browser, an Owner or Admin picks the Pingstack account, and you approve the connection. Clients register themselves, so there's no client ID or secret to copy.
Claude Code
claude mcp add --transport http pingstack https://app.pingstack.io/mcp
Then run /mcp, choose pingstack and sign in. Add --scope user to use it
in every project. You can also install it as a plugin, which adds the
skills below:
/plugin marketplace add atomicvan/pingstack-mcp
/plugin install pingstack@pingstack
Claude (claude.ai and Claude Desktop)
Go to Settings → Connectors, choose Add custom connector, name it
Pingstack and enter https://app.pingstack.io/mcp.
GitHub Copilot CLI
copilot mcp add --transport http pingstack https://app.pingstack.io/mcp
If /mcp shows needs-auth, run /mcp auth pingstack.
VS Code with Copilot
Run MCP: Add Server from the Command Palette, choose HTTP and enter
the URL. Or add it to .vscode/mcp.json:
{
"servers": {
"pingstack": {
"type": "http",
"url": "https://app.pingstack.io/mcp"
}
}
}
Cursor
Add this to ~/.cursor/mcp.json, then sign in from Cursor's MCP settings:
{
"mcpServers": {
"pingstack": {
"url": "https://app.pingstack.io/mcp"
}
}
}
OpenAI Codex
codex mcp add pingstack --url https://app.pingstack.io/mcp
codex mcp login pingstack
Google Antigravity
Install the plugin in the antigravity folder with the Antigravity CLI. The
app picks it up too. You sign in in the browser the first time it connects:
git clone https://github.com/atomicvan/pingstack-mcp
agy plugin install ./pingstack-mcp/antigravity
Or add this to ~/.gemini/config/mcp_config.json, then click Authenticate
next to pingstack under Agent settings → Customizations:
{
"mcpServers": {
"pingstack": {
"serverUrl": "https://app.pingstack.io/mcp"
}
}
}
Gemini CLI
Install it as an extension:
gemini extensions install https://github.com/atomicvan/pingstack-mcp
Then run /mcp auth pingstack.
The client setup guide covers each client in more detail, plus how to use an API key when your client can't sign in through a browser.
Tools
The server has 35 tools. Your client only sees the ones the connection has permission for.
Checks
| Tool | What it does |
|---|---|
get_account_capabilities |
Account, plan limits, usage, a count of checks by status, time zone and the fields each check type accepts |
list_checks |
List and filter checks, by one status or several |
get_check |
One check's settings, status, why it's down, last run and expiry date |
create_check |
Create a heartbeat, website, TCP, DNS, SSL certificate or domain check |
update_check |
Change a check's settings |
get_check_history |
Daily runs, latency and incidents for up to a year |
list_check_runs |
Individual runs for up to 90 days |
get_check_metrics |
Latency percentiles, by probe region or DNS resolver |
get_check_uptime |
Uptime over 7 days to a year |
Incidents
| Tool | What it does |
|---|---|
list_incidents |
Incidents with their cause, newest first, filtered by check and state |
get_incident |
One incident |
create_incident_update |
Post a status update on an incident |
Notification groups
| Tool | What it does |
|---|---|
list_notification_groups |
Alert groups, with their members and rosters |
list_account_members |
People on the account |
list_rosters |
On-call rosters |
create_notification_group |
Create a group from people and rosters already on the account |
update_notification_group |
Rename a group or change who it alerts |
assign_notification_group |
Add a group to many checks, or to every check on a dashboard |
Dashboards and status pages
| Tool | What it does |
|---|---|
list_dashboards |
Dashboards and status pages |
get_dashboard |
One dashboard, its sections and its checks |
create_dashboard |
Create a private dashboard, with checks already on it if you like |
update_dashboard |
Change a dashboard's name and display settings |
attach_check_to_dashboard |
Add one check or several to a dashboard |
detach_check_from_dashboard |
Remove a check from a dashboard |
reorder_dashboard_checks |
Reorder a dashboard's checks |
create_dashboard_section |
Add a section heading to a dashboard |
update_dashboard_section |
Rename a section or set which checks it holds |
publish_dashboard |
Make a dashboard a public status page (account Admins only) |
Off unless you turn them on
These permissions start unticked when you approve a connection, each with a Sensitive badge. Tick one only if your agent needs it.
| Tool | What it does |
|---|---|
get_check_ping_urls |
A heartbeat check's ping URLs, which work as credentials |
pause_check, resume_check |
Pause a check and resume it. A paused check sends no alerts |
delete_check |
Delete a check and its history for good |
delete_dashboard |
Delete a private dashboard. Its checks are kept |
delete_notification_group |
Delete an alert group |
unpublish_dashboard |
Take a status page offline |
A connection can delete 10 items a minute. A few minutes after it deletes anything, the account owner gets an email listing what was deleted.
History goes back as far as your plan keeps it. See the tool reference for every argument.
Skills
The Claude Code and Cursor plugins come with three skills. Each one walks your agent through a whole job, so you don't have to prompt every step.
| Skill | What it does |
|---|---|
monitor-cron-job |
Reads a job's schedule, creates a matching heartbeat check and adds the ping to the job |
client-dashboard |
Builds a dashboard for a client or team, organizes it into sections and sets up who gets alerted |
uptime-report |
Writes up uptime, incidents and their causes for a dashboard or the whole account |
In Claude Code, run one with /pingstack:monitor-cron-job, or describe what
you want and Claude picks the skill. The skills are plain Markdown in the
skills folder.
What it won't do
The server can't invite people, add notification channels such as email, Slack or SMS, edit rosters, change status page branding or domains, change a check's type or buy anything. Those stay in the Pingstack app. It never sends a test alert or runs a check on demand.
Every connection is limited to one account, to the permissions approved when it was connected, and to the approving person's current role. At a plan limit it tells your agent the limit and returns an upgrade link for a person to open.
Example prompts
- Add a daily heartbeat for the backup job with a 15 minute grace period and put it on the Infrastructure dashboard.
- Which checks had incidents in the last 7 days, what caused them, and how long were they down?
- Create a dashboard for the Northstar client with their checks in two sections, then a notification group with Alex and me in it, and alert that group for every check on the dashboard.
- Pause the staging checks while we migrate the database, and resume them when I say so.
- Create website and SSL certificate checks for every domain in this README.
- What's the p95 latency for the API check over the last 30 days?
- How many checks do I have left on my plan?
Support
- Docs: docs.pingstack.io
- Email: support@pingstack.io
- Privacy policy: pingstack.io/privacy-policy
- Terms: pingstack.io/terms-conditions
Please report security issues to support@pingstack.io rather than opening a public issue.
License
The files in this repository are under the MIT License. The Pingstack service is covered by its terms.
Perguntas sobre o MCP Pingstack
Como adiciono o MCP Pingstack ao Claude Code?
Execute claude mcp add --transport http pingstack-mcp https://app.pingstack.io/mcp no seu terminal. Adicione --scope user para torná-lo disponível em todos os projetos.
Como adiciono o MCP Pingstack ao Cursor?
Abra ~/.cursor/mcp.json e adicione o servidor em mcpServers, usando a configuração do Cursor mostrada na seção de instalação acima.
Quais ferramentas o MCP Pingstack oferece?
Ele expõe 33 ferramentas, incluindo get_account_capabilities, list_checks, get_check, create_check, update_check.
Quanto custa para usar?
O serviço é gratuito para até 15 verificações, e o MCP está incluso em todos os planos.
Preciso instalar algo localmente?
Não, o servidor roda no Pingstack, portanto não há nada para instalar ou executar em sua máquina local.
O agente pode enviar alertas de teste ou executar verificações sob demanda?
Não, o servidor não envia alertas de teste nem executa verificações sob demanda.