moscovium-mc/GitHub-Exploit-Tracker
Real-time GitHub exploit PoC monitor that scores, analyzes, and alerts on newly published CVE exploits before they trend. Built to give defenders a head start.
GitHub Early Warning Stalker Bot
Real-time GitHub exploit PoC monitor that scores, analyzes, and alerts on newly published CVE exploits before they trend. Built to give defenders a head start and red teams the edge.
What it does
- Scans GitHub every 5 minutes for new repos containing CVE identifiers (
CVE-2024,CVE-2025+ keywordspoc,exploit,bypass) - Threat analysis engine with red-flag detection (IPs, shellcode, base64 exec, obfuscation patterns)
- Automated confidence scoring (0–100) → SAFE / SUSPICIOUS / MALWARE verdicts
- Discord alerts with color-coded embeds, IOCs, author intel, and threat breakdowns
- SQLite deduplication — no spam, only new findings
- Deep README analysis — fetches and scans repository documentation for hidden indicators
- TypeScript with strict mode — zero runtime surprises, production-grade error handling
TL;DR: Your automated 0-day early warning system. GitHub publishes a PoC → you get alerted with threat analysis → you act before it goes mainstream.
Detection Capabilities
Red Flags Tracked
The bot analyzes each repository using a multi-point detection system:
| Threat Indicator | What It Detects | Score Impact |
|---|---|---|
| New GitHub accounts | Accounts created <7 days ago | -30 points |
| Suspicious fork/star ratio | High forks (>50) but low stars (<5) | -40 points |
| Embedded IP addresses | Hardcoded IPs in code/docs (C2 indicators) | -25 points |
| Base64 execution | base64, b64decode, eval(, exec(, atob( |
-25 points |
| PowerShell obfuscation | -enc, -encodedcommand, -w hidden |
-25 points |
| Shellcode patterns | Hex-encoded payloads (\x41\x42...) |
-25 points |
| Reverse shell indicators | socket.socket, connect(, sh -i, /bin/sh |
-25 points |
Scoring Logic
- Starting score: 70 points
- 60–100 → 🟢 SAFE (Likely legitimate security research)
- 30–59 → 🟡 SUSPICIOUS (Contains concerning patterns, needs review)
- 0–29 → 🔴 MALWARE (High-confidence malicious code)
Multiple red flags compound — a repo with shellcode + obfuscation + new account drops fast.
Getting it running
Prerequisites
- Node.js 18+ (if you don't have this, get it from nodejs.org)
- Discord bot token (create bot)
- GitHub Personal Access Token (classic token, no scopes needed — generate here)
- Discord channel ID (enable Developer Mode in Discord → right-click channel → Copy ID)
💡 Bot permissions needed: Send Messages, Embed Links, View Channel
Installation
# Clone the repository
git clone https://github.com/moscovium-mc/GitHub-Exploit-Tracker.git
cd GitHub-Exploit-Tracker
# Install dependencies
npm install
# Build the project
npm run build
# Run the bot
npm startConfiguration
Create a .env file in the project root:
# Discord Bot Token (from discord.com/developers/applications)
DISCORD_TOKEN=_bot_token
# Discord Channel ID (right-click channel → Copy ID)
DISCORD_CHANNEL_ID=channel_id
# GitHub Personal Access Token (github.com/settings/tokens)
GITHUB_TOKEN=github_token
# Scan interval in minutes (default: 5)
SCAN_INTERVAL_MINUTES=5Setting up Discord Bot
- Go to Discord Developer Portal
- Click New Application → give it a name
- Go to Bot section → Add Bot
- Copy the bot token (put in
DISCORD_TOKEN) - Under OAuth2 → URL Generator:
- Scopes:
bot - Permissions: Send Messages, Embed Links, View Channel
- Scopes:
- Use generated URL to invite bot to your server
Getting GitHub Token
- Go to GitHub Settings → Tokens
- Click Generate new token (classic)
- Name it (e.g., "Exploit Tracker Bot")
- No scopes needed for public repos
- Click Generate and copy the token
Platform Support
Fully Tested On
- Windows 10/11
- Linux (Ubuntu, Debian, Kali,)
- macOS (Intel & Apple Silicon)
Requirements
- Node.js 18.x or higher
- npm or yarn
- Internet access (GitHub API + Discord)
Troubleshooting
Bot starts but no alerts appear
Problem: Bot runs successfully but Discord channel stays empty
Solutions:
- Verify
DISCORD_CHANNEL_IDis the channel ID, not server ID - Confirm bot has Send Messages and Embed Links permissions
- Check bot can View Channel (role permissions)
- Look for errors in console output
"Could not fetch user data" warnings
Problem: Warnings about missing user metadata
Solutions:
- This is normal for some GitHub users (privacy settings)
- Bot continues scanning — doesn't break functionality
- User creation date check skipped for these accounts
Rate limit errors
Problem: GitHub API rate limit messages
Solutions:
- Authenticated requests: 5000/hour
- This bot uses ~60-80 per scan
- At 5-minute intervals = safe limits
- If hitting limits, increase
SCAN_INTERVAL_MINUTES
Database locked errors
Problem: SQLite "database is locked"
Solutions:
- Close any DB browser tools
- Delete
data/seen_repos.dband restart - Make sure only one bot instance running
No new repos found
Problem: Bot reports 0 repositories every scan
Solutions:
- GitHub API may be rate limiting
- Try increasing scan interval
- Verify
GITHUB_TOKENis valid and not expired - Check GitHub API status: githubstatus.com
Fully Working Features
- Real-time GitHub monitoring (CVE-2024/2025)
- Multi-query search (6 different query combinations)
- README content analysis
- 7-point threat detection system
- Automated confidence scoring
- IOC extraction (IPs, patterns)
- SQLite deduplication
- Discord embeds with color-coding
- Author intelligence (account age, profile)
- Configurable scan intervals
- Cross-platform support
- Error and retry logic
- Comprehensive logging
Use Cases
This bot is built for:
SOC Analysts
- Early visibility into public exploit drops
- Threat scoring reduces triage time
- IOC extraction for correlation
Threat Hunters
- Proactive intel on emerging TTPs
- Track exploit development trends
- Identify malicious repositories masquerading as PoCs
Red Teams
- Discover new PoCs before defenders patch
- Evaluate exploit reliability via threat score
- Monitor adversary tooling development
Bug Bounty Hunters
- Track exploit availability for in-scope CVEs
- Early PoC access for vulnerability research
- Correlate CVEs with public exploits
Incident Responders
- Match attacks to public exploits
- Identify compromised systems faster
- Understand attacker tooling
Security Researchers
- Monitor CVE weaponization timeline
- Track malware distribution via fake PoCs
- Research exploit development lifecycle
Performance Metrics
- Startup time: 2-3 seconds
- Scan duration: 30-60 seconds (depends on results)
- Memory usage: ~50-100 MB
- CPU usage: <5% during scans
- Network: ~5 MB per scan cycle
- Database size: ~1 KB per 100 repos tracked
Contributing
PRs welcome — but keep it lean and effective.
Good Additions
- New red-flag patterns (domain extraction, hash detection)
- Additional CVE years (CVE-2026+)
- Better IOC parsing (URLs, file hashes, domains)
- Severity-based filtering ("only alert score <40")
- Custom search queries via config
- Webhook support for non-Discord integrations
Please Don't Add
- Bloat or unnecessary dependencies
- Web dashboards (this is a bot, not a platform)
- Breaking changes without discussion
- Features that slow down scans
Contribution Guidelines
- Fork the repository
- Create a feature branch
- Test on Windows, Linux, and macOS
- Keep TypeScript strict mode compliance
- Update README if adding features
- Submit PR with clear description
Support My Work
If this bot saves you time (or prevents a breach), consider supporting development:
Crypto donations:
bc1quavqz6cxqzfy4qtvq4zxc4fjgap3s7cmxja0k40x5287af72afbc152b09b3bf20af3693157db9e425HYZjfEx8NbEMJX1vL1GmGj39zA6TgMsHm5KCHWSZxF4j86zv6vTDuG35sdBzBpwVAsD71hbt2gjH14qiesyrSsMkUAWHQkPZyY9TreeQ5dXRuP57yitP4Yn13SQEcMK4MhtwFzPoRR1
Your support keeps my projects maintained and improved.
Acknowledgments
- GitHub API - For providing access to repository data
- Discord.js - Discord bot framework
- Octokit - GitHub API client library
- Security research community - For responsible disclosure practices
Legal Disclaimer
This tool is designed for defensive security operations and authorized security research only.
Intended Use
- Monitoring public exploit repositories
- Threat intelligence gathering
- Security operations center (SOC) automation
- Vulnerability research and analysis
Prohibited Use
- Deploying exploits without authorization
- Attacking systems you don't own
- Violating GitHub's Terms of Service
- Any illegal activities
By using this software, you agree to use it responsibly and ethically. The authors are not responsible for misuse.
License
MIT License - See LICENSE file for details.
GitHub publishes → bot analyzes → you respond.
⭐ Star this repo if it helps your security operations.
Report issues to improve detection accuracy.
Fork and contribute to make it better.
Stay ahead. Stay secure.