| name | ai-traffic-detector |
| description | Detects AI coding agent traffic in server logs and analytics. Identifies Claude Code, Cursor, Cline, Aider, and other agent fingerprints. Use when analyzing AI agent access patterns to documentation. |
AI Traffic Detector Skill
Detects and analyzes AI coding agent traffic.
When to Use This Skill
- Identify AI agents accessing docs
- Set up AI traffic segmentation
- Monitor crawler access
- Analyze behavior patterns
Known Agent Fingerprints
| Agent | Pattern | Reliability |
|---|
| Aider | Aider/*, aider.chat | High |
| Claude Code | claude-cli/*, axios/1.8.x | Medium |
| Cursor | got + x-cursor-client-* | High |
| Windsurf | colly | High |
| OpenCode | Electron/* + OpenCode/* | High |
| Codex CLI | codex_cli_rs/* | High |
Accept Header Signal
AI agents send Accept: text/markdown - no human browser does this.
Usage
python3 scripts/ai_traffic_detector.py --log-file /var/log/nginx/access.log
python3 scripts/ai_traffic_detector.py --url https://example.com/docs