OpenCollab MCP
Provides tools for analyzing GitHub profiles (languages, topics, contribution patterns), finding beginner-friendly issues ('good first issue', 'help wanted') matched to user skills, scoring repository health and contributor-friendliness, assessing contribution readiness (setup difficulty, documentation, CI), generating PR plans from issue context, and estimating contribution impact (stars, reach, resume value) across open source projects.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@OpenCollab MCPanalyze my GitHub profile and suggest good first issues"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
π OpenCollab MCP
Land your first open source PR this weekend.
Stop scrolling GitHub. Let AI find you a mergeable issue in 30 seconds β matched to your actual skills, in a repo that's actually alive.
Works with Claude Desktop Β· Cursor Β· VS Code Β· any MCP client
Install in 60 seconds Β· See it in action Β· All 22 tools
The problem
You want to contribute to open source. So you:
Scroll through hundreds of GitHub repos π©
Find a "good first issue" β already taken π€
Spend an hour understanding a dead repo π
Discover someone already opened a PR π
Give up, go back to tutorials πΊ
This loop is broken. OpenCollab fixes it in one sentence to your AI assistant.
The fix
"Find me a good first issue I can contribute to this weekend."Claude calls OpenCollab β scans your GitHub profile β picks your strongest language β finds beginner-friendly issues in active repos with no existing PR β hands you the issue + full context to draft the fix.
One sentence. 30 seconds. A real mergeable issue.
π¦ Install in 60 seconds
1. Get a free GitHub token
github.com/settings/tokens β Generate new token (classic) β check public_repo β copy.
2. Add to your AI tool
Edit your config file:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.json
Add this:
{
"mcpServers": {
"opencollab": {
"command": "uvx",
"args": ["opencollab-mcp"],
"env": {
"GITHUB_TOKEN": "your_github_token_here"
}
}
}
}Restart Claude Desktop. Done.
Add to .cursor/mcp.json or your VS Code MCP config:
{
"mcpServers": {
"opencollab": {
"command": "uvx",
"args": ["opencollab-mcp"],
"env": {
"GITHUB_TOKEN": "your_github_token_here"
}
}
}
}pip install opencollab-mcpThen in your MCP config:
{
"mcpServers": {
"opencollab": {
"command": "opencollab-mcp",
"env": {
"GITHUB_TOKEN": "your_github_token_here"
}
}
}
}π¬ See it in action
The killer demo β 3 prompts to go from zero to a drafted PR
1οΈβ£ Analyze me
"My GitHub username is
prakhar1605. Am I ready to contribute to open source?"Readiness: 72/100. You know 4 languages, 15 public repos, haven't opened PRs yet. Tips: start with a docs fix, try a repo you already use.
2οΈβ£ Find me a mergeable issue
"Find me a Python good-first-issue I can finish in 1β2 hours. Make sure nobody's working on it."
Returns 5 issues Β· filters out ones with assignees or linked PRs Β· sorts by "quickness score" (short body, few comments, easy label).
3οΈβ£ Plan the PR
"Plan a PR for issue #456 in
owner/repo."Pulls the issue body, comments, CONTRIBUTING.md, the repo's directory structure, and the default branch β hands Claude everything needed to draft the actual code.
That's the whole loop: Analyze β Find β Plan β Ship.
More things you can just say
You say⦠| What happens |
"Is issue #123 in facebook/react still available?" | β No assignees, no open PRs. 3 comments, 12 days old. Go for it. |
"Compare langchain vs llama_index for contributing." | Side-by-side: stars, PR merge rate, activity. Recommends winner. |
"Is tensorflow/tensorflow alive?" | 847 commits in last 30 days. Growing +23%. Safe to invest time. |
"How complex is issue #5432 in pytorch?" | 7/10 Β· Advanced. 12 comments, architecture label. Skip unless you know the codebase. |
"Find Python repos with GSoC or Hacktoberfest." | Mentored repos sorted by mentor signals. |
"What dependencies does fastapi use?" | Reads pyproject.toml β starlette, pydantic, uvicorn. |
"What's the impact of contributing to react?" | π― MASSIVE Β· 230k+ stars Β· Resume line: "Contributed to a project used by millions of devs." |
π οΈ All 22 tools
Tool | Does |
| All-in-one β profile analysis + matched issues |
| Good-first-issues for any language |
| Trending repos seeking contributors |
| Find repos like one you already like |
| GSoC Β· Hacktoberfest Β· Outreachy repos |
| 1β2 hour issues β docs, typos, tests |
Tool | Does |
| Health score 0β100 |
| Setup difficulty (Dockerfile, CI, docs) |
| Impact tier + resume line |
| 30-day momentum β growing? dying? |
| Two repos side-by-side + winner |
| Language % breakdown |
| Tech stack β what libs the project uses |
Tool | Does |
| Deep profile analysis |
| Open source readiness 0β100 + tips |
| Top contributors of any repo |
Tool | Does |
| Is this issue still free? |
| Difficulty 1β10 |
| Old unclaimed issues β hidden wins |
| All labels + beginner-friendly ones |
| Recently merged PRs β what gets accepted |
| Full context for PR planning |
β‘ Why it's different
You ask Claude β Claude calls OpenCollab tools β Tools hit GitHub's free API β Data flows back β Claude reasons over it β You get a real, specific answerOpenCollab is a data bridge, not an AI. Your AI assistant does the thinking. That means:
π Zero AI costs β pure GitHub API, no paid services
π No secrets besides a free GitHub token
π» Runs locally on your machine (STDIO transport)
π Private β your GitHub data never leaves your computer
β‘ Fast β direct API calls, no middleware, no telemetry
ποΈ Develop / Contribute
This project is itself a great first contribution target.
git clone https://github.com/prakhar1605/Opencollab-mcp.git
cd Opencollab-mcp
pip install -e .
export GITHUB_TOKEN="your_token_here"
python -m opencollab_mcp.server
# Or test with the MCP Inspector:
npx @modelcontextprotocol/inspector python -m opencollab_mcp.serverCheck open issues labelled good first issue.
πΊοΈ Roadmap
22 tools shipped
Published on PyPI β
uvx opencollab-mcpworks out of the boxIn-memory caching layer (fewer API calls, less rate-limit friction)
first_pr_generatorβ one-shot "find + plan + draft my first PR"track_my_prsβ dashboard of your open PRs with staleness nudgesskill_gapβ compare your skills vs a target repo's stackGitHub Actions CI
SSE/remote deployment
Got an idea? Open an issue β that's the fastest path in.
π License
MIT β see LICENSE.
Built with β€οΈ by Prakhar Pandey Β· IIT Guwahati
β Star this repo if OpenCollab helps you land a PR. β
It's the single biggest thing you can do to help other devs discover it.
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/prakhar1605/Opencollab-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server