Render MCP server

What the model can do

Render: workspaces, services with type, region, plan, repository and suspension, each service's deploys and events, and Postgres databases. Can deploy, roll back or cancel a deploy, restart, suspend, resume or scale a service, run or cancel a one-off job, trigger or cancel a cron job run, and purge a web service's cache. Never reads environment variables, connection strings or logs.

Reads

• render_list_workspacesRender workspaces (owners) the key reaches: id, name and whether a team or a user.
• render_list_servicesRender services: name, type (web, private, background worker, cron, static site), region, plan, public URL, repository and branch, auto-deploy…
• render_list_deploysDeploys of one Render service, newest first as Render returns them: status (live, build_failed, update_failed, canceled, deactivated, in progress…)…
• render_list_eventsEvents on one Render service: deploys starting and ending, builds, server failures, restarts, suspensions, autoscaling, with their time and details.
• render_list_postgresRender Postgres databases: name, plan, region, status, Postgres version, whether high availability is on, disk size, replicas…

Acts, when you ask

These change something in Render. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• render_trigger_deployDeploys a Render service now, replacing what is live once it succeeds.
• render_cancel_deployCancels an in-progress Render deploy; what was live stays live.
• render_rollback_deployRolls a Render service back to an earlier deploy, which goes live in place of the current one.
• render_restart_serviceRestarts a running Render service; it is briefly unavailable while it comes back.
• render_suspend_serviceSuspends a Render service, which stops it running and serving anyone until resumed (and stops its compute billing).
• render_resume_serviceResumes a suspended Render service, so it runs and is billed again.
• render_scale_serviceSets a Render service to a fixed number of instances, which changes what it costs.
• render_run_jobStarts a one-off job on a Render service: runs the given command in a new instance built from the service, with its environment.
• render_cancel_jobCancels a running one-off job on a Render service.
• render_run_cron_jobRuns a Render cron job now, outside its schedule, cancelling any run already in progress.
• render_cancel_cron_job_runCancels the run of a Render cron job that is in progress now.
• render_purge_cachePurges the edge cache of a Render web service (when caching is on), so the next requests go to the service.

Connect it

First connect Render on Dashboard → Connections, then add Memtro to your client once. The Render tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared Render connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › Why did the last deploy of the api service fail?
• › What commit is live on the web service right now?
• › Which Render services are suspended?
• › Which free Postgres databases expire this month?
• › Roll the api service back to yesterday's deploy

Questions

Is this an official Render MCP server?

No. It is Memtro's MCP server with Memtro's Render connector behind it: the connector reads Render through its API with the credentials you connect on Dashboard → Connections, and Render publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in Render?

Only when you ask, and only what it has been granted: deploy a Render service now (latest commit, a given commit or image); cancel an in-progress Render deploy; roll a Render service back to an earlier deploy; restart a Render service; suspend a Render service (takes it offline); resume a suspended Render service; scale a Render service to a fixed number of instances; run a one-off job (a command) on a Render service; cancel a running Render one-off job; trigger a Render cron job run now; cancel a running Render cron job run; purge a Render web service's edge cache. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Engineering and operations

• GitHub MCP server
• AWS CloudWatch MCP server
• Sentry MCP server
• ServiceNow MCP server
• Okta MCP server
• Twilio MCP server