PD-Hunter

README

PD-Hunter

Open Source Bounty Intelligence Platform

Find high-value open source bounties matched to your skills, powered by AI.

[English] | 简体中文

---

Features

• 15+ Organizations — Tracks bounties across projectdiscovery, supabase, cal-com, appwrite, and more
• AI-Powered Analysis — GPT-4o generates friction level, technical hints, and bounty tier for each issue
• Hunter Cards — Beautiful hacker-themed dashboard with filtering, sorting, and search
• S-Tier Highlighting — High-value ($1000+) bounties prominently featured with glow effects
• Hidden Gems Detection — Find low-competition opportunities with customizable thresholds
• Expert Hint Preservation — Manual expert hints preserved across automated updates
• Auto-Update — GitHub Actions refreshes data every 6 hours
• Full-Text Search — Search bounties by title, repo, label, or technical hint

Quick Start

Option 1: CLI (Go)

# Install
go install github.com/FuZoe/PD-Hunter/cmd/hunter@latest

# Scan all organizations
export GITHUB_TOKEN=your_token
hunter scan --config mapping.json --output bounty_issues.json

Option 2: Full Pipeline

# 1. Clone
git clone https://github.com/FuZoe/PD-Hunter.git
cd PD-Hunter

# 2. Scan bounty issues
export GITHUB_TOKEN=your_token
go run ./cmd/hunter scan

# 3. Enrich with AI
pip install -r requirements.txt
python enrich_bounties.py

# 4. Start dashboard
cd frontend && npm install && npm run dev
# Open http://localhost:3000

Architecture

flowchart LR
    A[mapping.json<br/>15+ Orgs] -->|config| B[cmd/hunter scan<br/>Go CLI + GitHub API]
    B -->|bounty_issues.json| C[enrich_bounties.py<br/>GPT-4o Analysis]
    C -->|enriched_bounties.json| D[frontend/<br/>Next.js Dashboard]
    E[GitHub Actions<br/>Every 6 hours] -.->|triggers| B
    E -.->|triggers| C
    E -.->|deploys| D

Pipeline Stages

1. Scan — The Go CLI (cmd/hunter scan) reads mapping.json for target organizations and bounty labels, queries the GitHub Search API, counts open PRs per issue, and deduplicates results into bounty_issues.json.

2. Enrich — The Python script (enrich_bounties.py) feeds each issue to GPT-4o via GitHub Models to produce Hunter Intelligence: friction level, technical hint, bounty tier (S/A/B), and Hidden Gem flag. Expert hints are preserved across runs.

3. Publish — GitHub Actions runs the pipeline every 6 hours with data validation and failure alerting. The Next.js dashboard loads the enriched JSON and renders it as a filterable, searchable, hacker-themed card view.

Project Structure

PD-Hunter/
├── cmd/hunter/          # Go CLI entry point (cobra)
├── pkg/
│   ├── scraper/         # GitHub API client, config loader, types
│   └── exporter/        # JSON export
├── frontend/            # Next.js 14 + Tailwind + shadcn/ui
│   ├── src/app/         # Pages (dashboard)
│   ├── src/components/  # BountyCard, FilterBar, StatsPanel
│   ├── src/hooks/       # useBounties, useFilters
│   └── src/lib/         # Types, utilities, API
├── enrich_bounties.py   # AI enrichment (GPT-4o)
├── mapping.json         # Organization tracking config
├── static/              # Legacy static dashboard
└── .github/workflows/   # CI + auto-update pipelines

Tech Stack

Layer	Technology
CLI	Go 1.22 + cobra
AI Analysis	Python + OpenAI (GPT-4o via GitHub Models)
Frontend	Next.js 14 + Tailwind CSS + Lucide Icons
CI/CD	GitHub Actions (lint + test + build + deploy)
Testing	Go test (88% coverage) + ESLint + TypeScript

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

Common ways to contribute:

• Add organizations — Edit mapping.json and submit a PR
• Report bugs — Use the Bug Report template
• Request features — Use the Feature Request template

License

MIT