Skip to content
MCP server

gitgoblin

Find open-source projects worth contributing to and the work they need

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
0.1.8
Listing updated
3 days ago

About gitgoblin

gitgoblin is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.

The server is the read-only part of a larger project that helps people contribute to other people's software. It ranks public repositories against a goal you choose (contribute, learn, adopt, revive or explore) and shows the reasons for each rank, so a quiet library can outrank a crowded one. The pull-request step in the main project is still experimental, and nothing is published without a person approving it.

Four tools cover the flow: discover repositories, explain a score with its eleven signals, analyse a repository's profile and ranked opportunities, and fetch one opportunity with clone steps, commands and conventions. Analysis supports GitHub, GitLab and Codeberg; discovery reads GitHub.

The README treats repository text as hostile. It only reads repositories confirmed public, runs analysis in a child process with an empty home directory, stops after five minutes, allows two at once and strips control characters from output. A GitHub token is optional and lifts the rate limit from 60 to 5000 requests an hour.

Install

Run in a terminal

claude mcp add gitgoblin --env GITHUB_TOKEN='<GITHUB_TOKEN>' -- npx -y @gitgoblin/mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "gitgoblin": {
      "command": "npx",
      "args": [
        "-y",
        "@gitgoblin/mcp"
      ],
      "env": {
        "GITHUB_TOKEN": "<GITHUB_TOKEN>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "gitgoblin": {
      "command": "npx",
      "args": [
        "-y",
        "@gitgoblin/mcp"
      ],
      "env": {
        "GITHUB_TOKEN": "<GITHUB_TOKEN>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "gitgoblin": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@gitgoblin/mcp"
      ],
      "env": {
        "GITHUB_TOKEN": "<GITHUB_TOKEN>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "gitgoblin": {
      "command": "npx",
      "args": [
        "-y",
        "@gitgoblin/mcp"
      ],
      "env": {
        "GITHUB_TOKEN": "<GITHUB_TOKEN>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "gitgoblin": {
      "command": "npx",
      "args": [
        "-y",
        "@gitgoblin/mcp"
      ],
      "env": {
        "GITHUB_TOKEN": "<GITHUB_TOKEN>"
      }
    }
  }
}

Configuration

  • GITHUB_TOKENOptional GitHub token that can read public repositories. Raises rate limits from 60 to 5000 requests per hour and unlocks starter-issue data. A GitHub CLI login works too.
  • GITHUB_TOKENOptional GitHub token that can read public repositories. Raises rate limits from 60 to 5000 requests per hour and unlocks starter-issue data. A GitHub CLI login works too.

Tools this server provides (4)

Taken from the project documentation; the running server may differ.

discover_repositories
Rank public repositories for a goal with reasons for and against.
explain_repository
Show all eleven discovery signals with value, weight and evidence.
analyze_repository
Profile a repository and rank its opportunities with evidence.
get_opportunity
Full detail on one opportunity plus how to work on the project.

What it does

Goal-based ranking
Rank repositories for contributing, learning, adopting, reviving or exploring.
Score explanations
See each signal's value, weight and evidence.
Opportunity finding
Surface issues, stale pull requests and other checkable work.
Contribution briefing
Get clone, commands, conventions and etiquette for one task.
Hardened analysis
Runs repository analysis in an isolated child process with a timeout.
Opt-out respect
Projects whose maintainers opted out are left out.

Good for

  • Find a small project to contribute a first pull request to
  • Check whether a dependency's maintainers welcome outside contributors
  • Pick a revival candidate among quiet repositories
  • Brief an agent on one issue before it starts work

Pricing

Open source

MIT licensed; free

Prices change. Check anandkaranubc.github.io before you buy.

gitgoblin FAQ

Does it open pull requests?

No. The four tools are read-only, and the server instructions say it never opens issues, pull requests or comments unless asked.

Do I need a GitHub token?

Optional. It raises limits from 60 to 5000 requests an hour and unlocks starter-issue data.

Which hosts can it analyse?

analyze_repository and get_opportunity accept GitHub, GitLab and Codeberg; other hosts are refused.

How do I install it?

Run claude mcp add gitgoblin -- npx -y @gitgoblin/mcp, or use the .mcpb bundle for Claude Desktop.

Is gitgoblin free?

Yes. gitgoblin is open source, so you can use and self-host it for free.

How do I add gitgoblin to Claude Code?

Run this command in your terminal: claude mcp add gitgoblin --env GITHUB_TOKEN='<GITHUB_TOKEN>' -- npx -y @gitgoblin/mcp

What are the best alternatives to gitgoblin?

The closest alternatives to gitgoblin are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to gitgoblin

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools