MantisBT MCP Server

What is MantisBT MCP Server?

MantisBT MCP Server connects your Mantis/MantisBT bug tracker to Claude Code and other MCP-compatible AI clients via the Model Context Protocol. It lets developers read, create, and update issues, notes, attachments, tags, and relationships directly from their AI assistant, eliminating the need to switch between editor and browser.

How to use MantisBT MCP Server?

Zero‑install setup via npx @dpesch/mantisbt-mcp-server — just configure the environment variables MANTIS_BASE_URL and MANTIS_API_KEY. The server supports stdio (default) and HTTP (Streamable HTTP) transports, with optional Bearer token authentication for remote or Docker deployments.

Key features of MantisBT MCP Server

• Full issue lifecycle: create, read, update, delete issues and their notes, attachments, tags, and relationships
• Browse project metadata: list projects, versions, categories, and members
• Access saved filters, user profiles, and enum lookups (severity, status, priority, …)
• Optional semantic search: find issues by meaning, not just keywords, using a local offline embedding model (~80 MB, downloaded once)

Use cases of MantisBT MCP Server

• Manage bugs and feature requests without leaving your code editor
• Quickly search and filter issues across projects using saved filters or semantic meaning
• Add notes, attach files, and link related issues directly from an AI chat interface
• Retrieve project metadata (versions, categories, members) during planning or triage

FAQ from MantisBT MCP Server

What transports does the server support?

The server supports stdio (default) and HTTP (Streamable HTTP) transports, with optional Bearer token authentication for remote or Docker setups.

Does the server require a specific MantisBT version?

Yes, it requires MantisBT 2.23 or later with the REST API enabled.

How does semantic search work and what are its requirements?

Semantic search runs fully offline using a local embedding model that is downloaded once (~80 MB). It finds issues by meaning rather than exact keyword matches.

How do I install and configure the server?

Use npx @dpesch/mantisbt-mcp-server with no installation step. Set the MANTIS_BASE_URL and MANTIS_API_KEY environment variables.

Where do the issue data and embeddings live?

Issue data comes from your MantisBT instance via its REST API. The semantic search embedding model is stored locally on the machine running the server.