gitwho

GitHub OSINT & Profile Intelligence Tool

python3 gitwho.py

          _ __          __
   ____ _(_) /__       / /_  ____
  / __ `/ / __/ | /| / / __ \/ __ \
 / /_/ / / /_ | |/ |/ / / / / /_/ /
 \__, /_/\__/ |__/|__/_/ /_/\____/
/____/

GitHub OSINT & Profile Intelligence

usage: gitwho.py [-h] [-w] [--port PORT] [-v] [-l] [--json] [username ...]

Comprehensive GitHub profile intelligence tool that collects and analyzes public GitHub data. Works as both a CLI tool and a web application.

Try it now: sholuv.net/gitwho

Screenshots

Features

• Profile Intelligence: Bio, location, company, followers, following, account age
• Repository Analysis: All repos sorted by stars with language breakdown
• Pull Request Tracking: PRs to other projects — merged, pending, and rejected
• Interest Profiling: Rule-based analysis of focus areas from repos, starred repos, and languages
• Community Metrics: Follower ratio, PR merge rate, star count, org memberships
• Achievements: Scraped from GitHub profile (graceful fallback)
• Web Interface: Google-like search page with dark/light mode
• JSON Output: Machine-readable output for scripting
• Optional LLM Enhancement: Ollama (local) or Hugging Face (cloud) for richer summaries

Installation

git clone https://github.com/sho-luv/gitwho.git
cd gitwho
pip install -r requirements.txt

Usage

CLI

# Analyze a user
python3 gitwho.py sho-luv

# Analyze from GitHub URL
python3 gitwho.py https://github.com/torvalds

# Multiple users
python3 gitwho.py sho-luv torvalds github

# Show all repos (including 0 stars)
python3 gitwho.py sho-luv -v

# Show URLs for PRs and repos
python3 gitwho.py sho-luv --links

# JSON output
python3 gitwho.py sho-luv --json

# From a file of usernames
python3 gitwho.py usernames.txt

Web Interface

# Launch web server
python3 gitwho.py --web

# Custom port
python3 gitwho.py --web --port 8080

Open http://localhost:5000 in your browser.

Environment Variables

Variable	Required	Description
GITHUB_TOKEN	No	GitHub personal access token. Increases rate limit from 60 to 5,000 requests/hour.
OLLAMA_URL	No	Ollama server URL (e.g., http://localhost:11434) for local LLM summaries
HF_API_TOKEN	No	Hugging Face API token for cloud LLM summaries
GITWHO_CACHE_TTL	No	Web cache TTL in seconds (default: 3600)

Getting a GitHub Token (Free)

1. Go to GitHub → Settings → Developer settings → Personal access tokens → Fine-grained tokens
2. Generate a new token with no special permissions (public data only)
3. export GITHUB_TOKEN=your_token_here

Data Collected

• Profile: Name, bio, location, company, blog, social links, account creation date
• Repositories: All public repos with stars, forks, language, topics, descriptions
• Languages: Aggregated byte-count breakdown across top 20 repos
• Pull Requests: To other projects — merged (accepted), open (pending), closed (rejected)
• Starred Repos: What the user follows/bookmarks
• Following/Followers: Full network lists
• Organizations: Org memberships
• Events: Recent public activity
• Achievements: Scraped from profile page (Arctic Code Vault, Pull Shark, etc.)

Cloudflare Worker Deployment (Optional)

The worker/ directory contains a Cloudflare Worker that powers the hosted web version at sholuv.net/gitwho. To deploy your own:

1. Install Wrangler:

   npm install -g wrangler

2. Set up the Worker config:

   cp worker/wrangler.toml.example worker/wrangler.toml

3. Create a KV namespace and update wrangler.toml with the ID:

   wrangler kv:namespace create GITWHO_CACHE

4. Set your GitHub token as a Worker secret:

   wrangler secret put GITHUB_TOKEN

5. Deploy:

   cd worker && wrangler deploy

6. Update docs/app.js to point API_URL to your deployed Worker URL, or set window.GITWHO_API_URL before the script loads.

Project Structure

gitwho/
├── gitwho.py          # CLI entry point
├── github_api.py      # GitHub API data fetching
├── analyzer.py        # Rule-based profile analysis + LLM plugins
├── scraper.py         # HTML scraping for achievements
├── requirements.txt
├── web/
│   ├── app.py         # Flask web application
│   ├── templates/
│   │   ├── index.html     # Search page
│   │   └── results.html   # Results page
│   └── static/
│       ├── style.css      # Dark/light theme
│       └── script.js      # Theme toggle, UI interactions
└── README.md

Requirements

• Python 3.7+
• requests
• rich
• flask
• beautifulsoup4
• aiohttp (optional, for LLM backends)