github-talent-mcp

MCP server that searches, scores, and ranks GitHub developers for technical recruiting.

Demo

https://github.com/user-attachments/assets/2dfd82b4-3eb5-4f2b-bc0a-2580b95043e4

Profile deep dive

Get the full developer profile and activity score for torvalds on GitHub

Claude calls get_developer_profile("torvalds") and returns:

Torvalds has zero recent GitHub activity because kernel development flows through mailing lists, not GitHub PRs. The reputation floor (293K followers) overrides the behavioral score and sets it to 150.

Repo contributor ranking

Get the top contributors to huggingface/transformers and rank them for a founding ML engineer role at an AI startup

Claude calls get_repo_contributors("huggingface/transformers") → rank_candidates on the top 24 contributors:

Combined score = activity × 0.4 + relevance × 0.6. Relevance is keyword overlap with the job description (ML, AI, startup, engineer, etc.).

Related MCP server: GitHub MCP Server Plus

Installation

1. Clone and install

git clone https://github.com/carolinacherry/github-talent-mcp.git
cd github-talent-mcp
python3 -m venv .venv && source .venv/bin/activate
pip install -e .

2. Create a GitHub personal access token

Go to github.com/settings/tokens and create a fine-grained or classic token with these scopes:

Create a .env file in the project root:

GITHUB_TOKEN=ghp_xxxxxxxxxxxx

3. Connect to Claude

Claude Code (CLI)

One command:

claude mcp add github-talent -- /path/to/github-talent-mcp/.venv/bin/python3 -m github_talent_mcp

Then set the token as an environment variable. Either:

• Export it in your shell: export GITHUB_TOKEN=ghp_xxxxxxxxxxxx

• Or keep it in the .env file — the server reads it via python-dotenv on startup

Restart Claude Code to pick up the new server. Verify with /mcp — you should see 8 tools under github-talent.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "github-talent": {
      "command": "/path/to/github-talent-mcp/.venv/bin/python3",
      "args": ["-m", "github_talent_mcp"],
      "cwd": "/path/to/github-talent-mcp",
      "env": {
        "GITHUB_TOKEN": "ghp_xxxxxxxxxxxx"
      }
    }
  }
}

Restart Claude Desktop. The tools will appear in the toolbox icon.

Try It

Once installed, paste these prompts to verify everything works:

Basic search:

Find Python developers in Raleigh active in the last 60 days

Profile deep dive:

Get the full developer profile and activity score for torvalds on GitHub

Full workflow:

Find 10 ML engineers in San Francisco active in the last 30 days, then rank them for a senior LLM inference engineer role

Repo contributors:

Get the top contributors to huggingface/transformers and rank them for a founding ML engineer role at an AI startup

JD scoring:

Score these candidates against this job description: [paste JD]. Candidates: tiangolo, karpathy, hwchase17

Compare candidates:

Compare tiangolo and hwchase17 for a Senior Python AI Engineer role

Bulk scoring:

Score these 10 GitHub usernames and give me a ranked table: [paste list]

Outreach:

Generate a casual recruiter message for tiangolo about a Senior Python role at Acme. My name is Daniel.

Tools

Scoring

The activity score combines two layers: behavioral signals (what you did recently) and a reputation floor (what you've built over time).

Behavioral Score (0-205)

Reputation Floor

The behavioral score alone penalizes developers whose work doesn't produce GitHub events — Torvalds works through mailing lists, senior maintainers merge via org bots, and many engineers work in private repos.

The reputation floor ensures cumulative impact isn't erased by a quiet quarter:

The final score is max(behavioral_score, reputation_floor). If the floor is applied, the breakdown includes a reputation_floor field so you know.

Score Tiers

• 150+ — exceptional (top OSS maintainers, well-known engineers)

• 120-149 — strong signal, worth reaching out

• 80-119 — solid developer with meaningful public work

• 40-79 — active but limited public signal

• <40 — low signal (likely private work or junior)

Ranking

rank_candidates combines the activity score with a relevance score (0-100) based on keyword overlap between the job description and the candidate's profile (bio, languages, repo topics, README). The combined score weights relevance at 60% and activity at 40% — a high-activity developer with no overlap to the job shouldn't outrank a relevant one.

Rate Limits

GitHub REST API: 5,000 requests/hour with token. A typical workflow (search + enrich 5 candidates + rank) uses ~60-100 API calls. Profile results are cached within a session to avoid redundant calls during ranking.

License

MIT