issue-tracker-mcp

# issue-tracker-mcp A minimal [Model Context Protocol](https://modelcontextprotocol.io) server that lets an AI assistant manage a kanban issue board. It uses the **streamable HTTP** transport, so the same code works for local development *and* for a hosted deployment that a whole team connects to. ## The exercise `src/server.ts` has the server scaffolding and **one** finished tool (`list_issues`) so you can see what a registration looks like. Four more tools are left as `TODO` comments — your job is to implement them with `server.registerTool(...)`: - `get_issue` - `create_issue` - `update_issue` - `delete_issue` The data layer in `src/store.ts` is already complete; you only need to wire it up. After each one, restart the server and try it in the Inspector. > Stuck? The `solution` branch has a fully working version. ## Run it ```bash npm install npm start # → Issue Tracker MCP server listening on http://localhost:3001/mcp ``` ## Poke it manually The MCP Inspector is a small web UI for calling tools by hand: ```bash npm run inspect # then choose "Streamable HTTP" and enter http://localhost:3001/mcp ``` ## Connect a client | Client | How | | --- | --- | | Claude Code | `claude mcp add --transport http issue-tracker http://localhost:3001/mcp` | | Claude Desktop / Claude.ai | Settings → Connectors → Add custom connector → paste the URL | ## Deploy for your team This is a plain Express app. Deploy `npm start` to any Node host (Render, Railway, Fly.io, Cloudflare Workers, a VPS) and share the resulting `https://…/mcp` URL — everyone who adds that URL talks to the same board.