Onplana MCP Server, connect any AI agent to your project portfolio

Claude Desktop

Open Settings → Developer → Edit Config. Paste this entry under mcpServers:

{
  "mcpServers": {
    "onplana": {
      "url": "https://mcp.onplana.com/mcp",
      "headers": {
        "Authorization": "Bearer pat_paste-your-token-here"
      }
    }
  }
}

Restart Claude Desktop. Ask "list my Onplana projects" to verify the connection.

ChatGPT (web + desktop)

ChatGPT registers MCP servers as custom connectors via the web UI, not a config file. Available on Plus, Pro, Team, Enterprise and Edu plans:

1. Open chatgpt.com → Settings → Connectors and click Add custom connector.
2. Name it Onplana, paste the URL https://mcp.onplana.com/mcp, choose Authentication: Bearer token, and paste your PAT.
3. Save. In a new chat, click the connector icon and pick Onplana, or ask "using Onplana, list my active projects".

On Plus/Pro: the connector is available in Deep Research and Agent mode. On Business plans, an admin may need to enable custom connectors at the workspace level first.

ChatGPT Codex CLI

Codex CLI reads ~/.codex/config.toml. Add the Onplana server under [mcp_servers.onplana]:

[mcp_servers.onplana]
url = "https://mcp.onplana.com/mcp"

[mcp_servers.onplana.headers]
Authorization = "Bearer pat_paste-your-token-here"

Start a fresh session with codex. Tools appear in the /mcp picker; ask the agent to call search_org_knowledge to confirm the wiring.

Gemini CLI + Gemini Code Assist

Both the gemini CLI and Gemini Code Assist (VS Code + JetBrains) read the same JSON config from ~/.gemini/settings.json (user-wide) or .gemini/settings.json (workspace). One snippet covers both:

{
  "mcpServers": {
    "onplana": {
      "httpUrl": "https://mcp.onplana.com/mcp",
      "headers": {
        "Authorization": "Bearer pat_paste-your-token-here"
      },
      "timeout": 5000
    }
  }
}

Note the key is httpUrl (not url) for remote HTTP MCP servers. Reload your editor window or restart the gemini CLI to pick up the change, then ask "list my Onplana projects via the onplana tool".

Gemini Enterprise (Vertex AI Agent Builder) uses a different OAuth-only admin flow via the Cloud Console. The consumer gemini.google.com app does not yet support custom MCP servers as of May 2026; use Gemini CLI or Code Assist instead.

Cursor

Add to ~/.cursor/mcp.json with the same JSON shape as Claude Desktop. Cursor picks up changes on its next launch:

{
  "mcpServers": {
    "onplana": {
      "url": "https://mcp.onplana.com/mcp",
      "headers": {
        "Authorization": "Bearer pat_paste-your-token-here"
      }
    }
  }
}

GitHub Copilot (VS Code)

VS Code's Copilot MCP support uses a servers key (not mcpServers). For a single repo, add .vscode/mcp.json at the workspace root:

{
  "servers": {
    "onplana": {
      "url": "https://mcp.onplana.com/mcp",
      "headers": {
        "Authorization": "Bearer pat_paste-your-token-here"
      }
    }
  }
}

Or wire it globally via Settings → Chat → MCP: Servers. Open Copilot Chat in Agent mode and the Onplana tools appear in the tool picker.

Custom MCP client

Any client supporting the MCP Streamable HTTP transport works. Use the connection details below; smoke-test with a tools/list JSON-RPC request:

# Smoke test, lists the curated tools
curl -X POST https://mcp.onplana.com/mcp \
  -H "Authorization: Bearer pat_paste-your-token-here" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

Onplana's open-source TypeScript SDK at github.com/Onplana/onplana-mcp-server wraps this for in-house agents.