Skip the install. Get this working in under 2 minutes.
Start a free trial on cloud.anythingmcp.com, add the Render in one click, then point your AI client (Claude, ChatGPT, Copilot or Cursor) at the generated MCP endpoint. No Docker, no git clone, zero engineering experience required.
AnythingMCP is listed in Claude's connector directory: add it to Claude in one click (opens in a new tab), then sign in and choose which MCP servers Claude may use.
Summary
Let your AI check and operate your Render services: services, deploys, logs, metrics, events, jobs, custom domains and Postgres and Key Value instances, and deploy, roll back, restart, suspend or resume services after you confirm.
Try asking
Example prompts for Render
Click any prompt to copy it. Paste into Claude, ChatGPT, Cursor, Gemini, Copilot or OpenClaw to run it against this connector.
Claude is AI and can make mistakes. Please double-check responses.
💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Render, paste your credentials, mint an MCP API key and you're done. No Docker, no
git clone, no local server.
Render + Claude
Let your AI check and operate your Render services: services, deploys, logs, metrics, events, jobs, custom domains and Postgres and Key Value instances, and deploy, roll back, restart, suspend or resume services after you confirm.
Prerequisites
See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:
RENDER_API_KEY
Step 1 — Get credentials
- Sign in to the Render Dashboard at
https://dashboard.render.com. - Open Account Settings → API Keys and click Create API Key. Give it a name and copy the key: it starts with
rnd_. - Paste it into
RENDER_API_KEYas Render shows it, without the word Bearer, and install the connector. The key is sent only to api.render.com. - Check the setup with
render_get_current_user, thenrender_list_workspaces.
Who sees what: a Render API key has your full rights in every workspace you belong to; Render has no read-only keys. Share the MCP server only with people who may deploy. For a read-only setup, switch off the write tools listed below.
Changes live services, so the AI asks first: render_trigger_deploy builds and deploys a new version (the latest commit of the branch, a specific commit or an image), which goes live when the build succeeds. render_rollback_deploy puts an earlier deploy live again; auto deploys stay on, so the next push to the branch can bring the change back. render_cancel_deploy stops a deploy that is still building or updating, and the current version keeps running. render_restart_service restarts all instances, so requests may fail for a moment. render_suspend_service stops the service: it serves no traffic until it is resumed. render_resume_service starts it again, and it is billed again. render_create_job runs a one-off command against your live environment, for example a database migration; it can change data and is billed per second. Before each of them the AI tells you the service and exactly what will happen, and waits for your yes.
Switched off at install: render_get_env_var_value, which returns the value of one environment variable in plain text (often a database URL or an API key). Turn it on only if the AI really needs a value; render_list_env_var_keys shows the variable names only. The Postgres and Key Value tools show status, plan, region, version and similar details, but never connection strings or passwords. Nothing in this connector can delete services, databases, domains or variables.
How to work: render_list_workspaces gives the workspace id (tea-... or usr-...) and render_list_services the service ids (srv-..., or crn-... for cron jobs); render_get_service shows branch, auto deploy, plan and build and start commands. For a failed deploy the AI uses render_list_deploys (filtered by status, for example build_failed), render_get_deploy and render_list_events. Logs come from render_list_logs with the workspace id as owner_id and the service id as resource; filter by type build for build output, by level error, by text or by HTTP status code (for example 5*). render_get_metrics reads CPU, memory, HTTP requests, HTTP latency, bandwidth, instance count, disk usage or active connections. Data stores: render_list_postgres, render_get_postgres, render_list_key_value and render_get_key_value. Domains and jobs: render_list_custom_domains, render_list_jobs and render_get_job (a job's output is in the logs, with the job id as resource). render_list_projects shows how services are grouped into projects and environments.
Times are ISO 8601 in UTC, for example 2026-10-15T08:00:00Z. Logs and metrics default to the last hour; when a log answer has hasMore, the AI continues with nextStartTime and nextEndTime.
Common errors: 401: the API key is wrong or was revoked; create a new one under Account Settings → API Keys. 403: the key works, but its user may not do this in that workspace. 404: a wrong id, or a resource in a workspace the key cannot reach. 429: the rate limit is reached; wait and retry.
The connector does not create, update or delete services, databases, environment groups, domains or environment variables, and it does not cover scaling, blueprints or webhooks.
Step 2 — Install the adapter
curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d
Step 3 — Add the connector in Claude
Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (
https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide showsYOUR_SERVER_ID.
Recommended path: works on claude.ai web without editing any config file.
- In Claude, open Customize → Connectors.
- Click +, then Add custom connector.
- Enter a name, e.g.
Render, and the remote MCP server URLhttps://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID, then click Add. - Click Connect and sign in to AnythingMCP (OAuth) to allow access.
All tools below appear in your chat: start typing prompts.
Available tools
| Tool | What it does |
|---|---|
render_get_current_user | Get the Render user the API key belongs to: id, name and email |
render_list_workspaces | List the workspaces (owners) the API key can reach: id, name, email, type (user or team) and whether two-factor authentication is on |
render_list_projects | List projects, which group services and databases into environments: id, name, workspace and environment ids |
render_list_services | List services: id, name, type (web_service, private_service, background_worker, static_site, cron_job), repository and branch, URL, region… |
render_get_service | Get one service: type, repository, branch, auto deploy, URL, region, plan, runtime, instances, health check path, build and start commands… |
render_list_deploys | List a service's deploys, newest first: id, status (live, build_in_progress, build_failed, update_failed, canceled, deactivated ...)… |
render_get_deploy | Get one deploy: status, trigger, commit id and message, image and start and finish time |
render_trigger_deploy | Start a new deploy of a service: the latest commit of its branch by default, a specific commit, or an image |
render_cancel_deploy | Cancel a deploy that is still building or updating; the service keeps running its current version |
render_rollback_deploy | Roll a service back to an earlier deploy: Render deploys that version again and it goes live |
render_restart_service | Restart all running instances of a service without a new deploy; requests may fail for a moment while instances restart |
render_suspend_service | Suspend a service: it stops running and stops serving traffic (and billing for compute) until resumed |
render_resume_service | Resume a suspended service: Render starts it again and it serves traffic and is billed |
render_list_env_var_keys | List the names of a service's environment variables |
render_get_env_var_value | Read the value of ONE environment variable of a service in plain text |
render_list_custom_domains | List a service's custom domains: id, name, type (apex or subdomain), verification status, redirect target and creation time |
render_list_jobs | List one-off jobs run on a service: id, start command, plan, status (pending, running, succeeded, failed, canceled) and times |
render_get_job | Get one job: start command, plan, status and times |
render_create_job | Run a one-off job: a command executed once in a new instance built from the service's latest successful build, with its environment, for… |
render_list_logs | Read logs of services, jobs, cron jobs, Postgres or Key Value instances in a time range (default the last hour), newest first by default… |
render_get_metrics | Read a metric as time series for a service, Postgres or Key Value instance over a time range (default the last hour): CPU, memory, HTTP… |
render_list_events | List a service's recent events: deploys started and ended, builds, restarts, suspensions, autoscaling, server failures and more, each with… |
render_list_postgres | List Render Postgres databases: id, name, status, plan, region, version, database name and user, disk size, high availability, read… |
render_get_postgres | Get one Render Postgres database: status, plan, region, version, database name and user, disk, high availability, replicas, pending… |
render_list_key_value | List Render Key Value (Redis-compatible) instances: id, name, status, plan, region, version, eviction and persistence options and dashboard… |
render_get_key_value | Get one Render Key Value instance: status, plan, region, version, options, IP allow list and pending maintenance |
Example prompts
- "Which of our Render services had a failed deploy this week? Show me the commit and the errors from the build log."
- "Our api service has been returning 5xx errors since 08:00 UTC. Check the request logs and the HTTP latency of the last two hours and tell me what changed."
- "Roll the checkout service back to the last deploy that was live before today. Show me the target deploy first and ask me before you roll back."
- "Run npm run migrate as a one-off job on the api service. Tell me the exact command and the service and wait for my yes."
- "Show status, plan and region of our Postgres and Key Value instances, and the CPU and memory of the web service over the last 24 hours."
FAQ
Does it work with Claude Code as well as Claude Desktop? Yes, point both at https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID.
Next steps
Was this guide helpful?
