MCP Integration

Connect external tools and APIs to List Coder using the Model Context Protocol.

MCP Integration

List Coder supports the Model Context Protocol (MCP) — an open standard that allows AI agents to connect to external tools, APIs, and data sources.

What is MCP?

MCP (Model Context Protocol) is a protocol that enables AI models to interact with external systems. Through MCP, List Coder's AI agents can:

  • Query databases
  • Call external APIs
  • Access file systems
  • Interact with development tools
  • Connect to custom services

Setting Up MCP

1. Install an MCP Server

MCP servers are lightweight programs that expose tools to List Coder. Install one using npm:

npm install -g @modelcontextprotocol/server-filesystem

2. Configure List Coder

Open Settings (Ctrl + ,) → AI → MCP Servers and add your server:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/workspace"]
    }
  }
}

3. Start Using

Once configured, the AI agent can automatically use MCP tools when relevant. You can also explicitly ask the agent to use a specific tool:

"Use the filesystem server to list all files in the src directory"

Available MCP Servers

Official Servers

ServerDescriptionInstall
FilesystemRead/write files on your systemnpm i -g @modelcontextprotocol/server-filesystem
GitHubInteract with GitHub repos, PRs, and issuesnpm i -g @modelcontextprotocol/server-github
DatabaseQuery SQL databasesnpm i -g @modelcontextprotocol/server-postgres
Brave SearchWeb search capabilitiesnpm i -g @modelcontextprotocol/server-brave-search
PuppeteerBrowser automationnpm i -g @modelcontextprotocol/server-puppeteer

Community Servers

ServerDescription
SlackSend messages and manage channels
JiraCreate and manage tickets
AWSInteract with AWS services
DockerManage Docker containers
KubernetesManage K8s clusters

Creating Custom MCP Servers

You can build your own MCP server to expose custom tools:

Example: Custom API Server

import { McpServer } from '@modelcontextprotocol/server';

const server = new McpServer({
  name: 'my-api',
  version: '1.0.0',
});

// Define a tool
server.tool(
  'get-user',
  'Fetch a user by ID',
  { userId: { type: 'string' } },
  async ({ userId }) => {
    const response = await fetch(`https://api.example.com/users/${userId}`);
    const data = await response.json();
    return { content: [{ type: 'text', text: JSON.stringify(data) }] };
  }
);

server.listen();

Publishing Your Server

  1. Publish to npm: npm publish
  2. Share with the community
  3. Submit to the MCP server registry

MCP in AI Agent

When MCP is configured, the AI Agent can:

  1. Discover available tools — Automatically detects configured MCP servers
  2. Select the right tool — Chooses the appropriate tool based on your request
  3. Execute operations — Calls the tool and returns results
  4. Handle errors — Gracefully manages failures and retries

Example Workflow

You: "Check if there are any open issues on my GitHub repo and create a summary"

Agent:
1. Uses GitHub MCP server to fetch issues
2. Analyzes the issues
3. Creates a summary document
4. Offers to create a report file

Security Considerations

  • MCP servers run locally — no data is sent to external services unless you configure it
  • Each server has access only to the resources you explicitly grant
  • You can review and approve tool calls before execution

Permissions:

Some MCP servers require additional permissions (e.g., filesystem access). Always review what permissions you're granting before enabling a server.

Troubleshooting

Server not connecting

  • Verify the server command and arguments are correct
  • Check that the server package is installed (npm list -g)
  • Look at the Output panel for error messages

Tools not appearing

  • Ensure the server is running (check the MCP status indicator)
  • Verify the server's tool definitions match the expected format
  • Restart List Coder after configuration changes