Skip to main content
Glama

Coda MCP Server (Koyeb + LangSmith)

This repo runs a Coda MCP server with:

  • Streamable HTTP endpoint: /mcp (recommended)

  • Legacy SSE endpoint: /sse

Quick Start (Koyeb First)

1. Fork or use this repo on GitHub

Use this repository in your own GitHub account (for example RunLittleTurtle/mcp-coda).

2. Create a Koyeb account and connect GitHub

In Koyeb:

  1. Create a new Web service

  2. Select your GitHub repo and main branch

  3. Keep Buildpack as builder

  4. Choose Free / CPU Eco instance

3. Set environment variables

Add these variables in Koyeb:

  • API_KEY = your Coda API key

  • PORT = 3000

Koyeb environment variables

4. Configure port

In Koyeb Ports:

  • Port: 3000

  • Protocol: HTTP

  • Public HTTPS access: enabled

Koyeb port configuration

5. Deploy

Click Save and deploy.

After deploy, your public URL will look like:

  • https://your-service-name.koyeb.app

Health check:

  • https://your-service-name.koyeb.app/health

MCP endpoint:

  • https://your-service-name.koyeb.app/mcp

LangSmith Agent Builder Setup

Add an MCP server with:

  • Name: mcp-coda (or your preferred name)

  • URL: https://your-service-name.koyeb.app/mcp

  • Authentication: Static Headers

  • Headers: none

LangSmith MCP settings

Important:

  • Use /mcp (not / and not /sse in Agent Builder)

  • If tools do not appear after a redeploy, wait a few seconds and reload Agent Builder

Local (minimal)

pnpm install pnpm build echo "API_KEY=your-coda-key" > .env PORT=3000 pnpm start:http

Then test:

  • http://localhost:3000/health

  • http://localhost:3000/mcp

Environment variables

  • API_KEY or CODA_API_KEY (required)

  • PORT (optional, default 3000)

Credits

License

MIT

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/RunLittleTurtle/mcp-coda'

If you have feedback or need assistance with the MCP directory API, please join our Discord server