Render MCP Server

Content

The Render MCP Server is the official Model Context Protocol server from render-oss that lets LLMs interact with your Render resources. Through it, an MCP-compatible client can create and manage web services, static sites, cron jobs, Postgres databases, and Key-Value instances; monitor deploys; query application logs and performance metrics; and run read-only SQL against Render-hosted Postgres.

It is written in Go and licensed under Apache-2.0. Render recommends connecting to the hosted server at https://mcp.render.com/mcp so you always run the latest version; a local binary (stdio or http transport) is also available for users who prefer to run it themselves. Authentication is via a RENDER_API_KEY scoped to your Render workspace.

Source Review

The following real source files were reviewed to verify the claims below:

• README.md — overview, use cases, and the full tool catalog with parameters
• main.go — entrypoint; --transport/-t flag (default stdio, also http) and --version/-v
• cmd/server.go — server bootstrap invoked by cmd.Serve(transport)
• go.mod — module path github.com/render-oss/render-mcp-server, Go 1.26
• SECURITY.md — vulnerability disclosure via HackerOne; 0.x supported
• LICENSE — Apache-2.0
• Official docs: render.com/docs/mcp-server — hosted vs. local install, config JSON, and RENDER_API_KEY

Repository signals at time of review: not archived, Go, Apache-2.0, 139 stars, latest release v0.3.0, last pushed 2026-06-04.

Features

Tools exposed by the server (grouped as in the README):

• Workspaces — list_workspaces, select_workspace, get_selected_workspace
• Services — list_services, get_service, create_web_service, create_static_site, create_cron_job, update_environment_variables
  • Runtimes: node, python, go, rust, ruby, elixir, docker
  • Regions: oregon, frankfurt, singapore, ohio, virginia
• Deployments — list_deploys, get_deploy
• Logs — list_logs, list_log_label_values (filter by resource, level, type, instance, host, status code, method, path, text, and time range)
• Metrics — get_metrics (CPU, memory, instance count, HTTP request count, HTTP latency, bandwidth, active connections; with aggregation, quantile, and host/path filters)
• Postgres — query_render_postgres (read-only SQL), list_postgres_instances, get_postgres, create_postgres
• Key-Value — list_key_value, get_key_value, create_key_value

Transports: stdio (default) and http, selectable with the --transport / -t flag on the local binary.

Installation

1. Create a Render API key from your Account Settings page (dashboard.render.com/u/settings).

2a. Hosted server (recommended). Point your MCP client at Render's hosted endpoint. For HTTP-capable clients such as Cursor (~/.cursor/mcp.json), pass the API key as a Render server-scoped authorization header:

{
  "mcpServers": {
    "render": {
      "url": "https://mcp.render.com/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_API_KEY>"
      }
    }
  }
}

For Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json), Render documents using mcp-remote. Keep RENDER_API_KEY inside the render server definition so other local MCP servers do not inherit it:

{
  "mcpServers": {
    "render": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.render.com/mcp",
        "--header",
        "Authorization: Bearer ${RENDER_API_KEY}"
      ],
      "env": {
        "RENDER_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

2b. Local binary (alternative). Download and unzip the executable for your architecture from the GitHub Releases page (an install script and build-from-source via Go are also documented), then configure your client to launch it over stdio:

{
  "mcpServers": {
    "render": {
      "command": "/path/to/render-mcp-server",
      "args": ["--transport", "stdio"],
      "env": {
        "RENDER_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Render notes: "We strongly recommend using Render's hosted MCP server instead of running it locally," because the hosted version updates automatically. See the official docs for the latest steps.

Use Cases

Per the project's own stated use cases:

• Creating and managing web services, static sites, cron jobs, and databases on Render
• Monitoring application logs and deployment status to help troubleshoot issues
• Monitoring service performance metrics for debugging, capacity planning, and optimization
• Querying your Postgres databases directly inside an LLM

Safety and Privacy

This server performs real, billable, write operations against your Render account. create_web_service, create_static_site, create_cron_job, create_postgres, and create_key_value provision live infrastructure, and update_environment_variables replaces the complete env-var set for a service — an incomplete array can drop existing variables. Created services run the build and start commands you provide, so treat generated commands as code execution on Render's platform. Review and confirm tool calls before approving them.

On privacy: the RENDER_API_KEY grants management access to your workspace(s). query_render_postgres returns row data, and the logs/metrics tools surface application log contents and performance data to the model — avoid exposing secrets you don't want the LLM to see. With the hosted option, requests transit Render's hosted MCP endpoint (https://mcp.render.com/mcp) rather than staying entirely local. Vulnerabilities should be reported via Render's HackerOne program (see SECURITY.md), not public GitHub issues.

Duplicate Check

No existing entry in the awesome-claude MCP directory matches this server. A search of the directory index for render-oss, render-mcp, and render mcp returned zero hits. The directory does contain official Netlify and Sentry MCP entries, but there is no Render MCP entry — this is not a duplicate.