MCP Overview

Kemon's MCP server lets AI coding assistants — like Claude Code, Cursor, and Codex — read and write your workspace data without leaving your editor.

What is MCP?

Model Context Protocol (MCP) is an open standard that defines how AI tools connect to external services. Instead of copy-pasting ticket IDs or switching to a browser, your AI assistant can interact with Kemon natively through a set of structured tools.

What you can do

Read data:

• List and search tickets by status
• Get full ticket details including description, assignees, and labels
• List teams, projects, labels, and workspace members

Write data:

• Create and update tickets with full metadata (status, priority, assignee, labels, project, due date)
• Add comments to tickets
• Attach files to tickets
• Batch-create tickets from codebase audit findings

How it works

Kemon exposes a dedicated MCP server at mcp.kemon.io. Each request is authenticated via a private API key passed in the Authorization: Bearer header. There are no persistent connections or sessions — the server handles each request independently, making it compatible with serverless environments and any MCP client that supports Streamable HTTP transport.

Scopes

API key scopes control which MCP tools are available. You select scopes when creating an API key.

Next steps

• Set up your MCP client — connect Claude Code, Cursor, Codex, and more
• Tools reference — full parameter and response documentation for all available tools