⬢github Python · 1.5K ★ +689 since we first saw it · pushed 16 h ago · MIT
N1nEmAn/sapper-v2
Autonomous pentest legion — AI commander + worker swarm + layered fact-intent graph. 3rd Tencent Intelligent Attack-Defense online champion (634 teams · 8h unattended · 1B tokens beat 4.5B). The Nüwa Test: give AIs Create/Language/History, let a civilization grow.
sapper-v2 is a Python multi-agent orchestration harness for autonomous penetration testing and CTF work. An AI commander dispatches worker agents that attack in parallel, with shared memory in a layered fact-intent graph, and agents/models are swappable via CLI flags or env vars.
Why now: It won the 3rd Tencent Intelligent Attack-Defense online qualifier (634 teams, 8h unattended) matching the runner-up's flags on ~1B tokens versus 4.5B, and it's trending on GitHub with 844 stars.
Who it is for: Security researchers and CTF competitors running authorized autonomous pentesting, plus builders of long-horizon multi-agent systems.
ai-agentsautonomous-agentsclaude-codectfllmmulti-agentpentest
Stars over our 146 snapshots: 836 to 1.5K, since 2 d ago.
Where people talked about it
- ⬢github new repos, most starred 6 min ago
- ⬢github new repos, most starred 18 h ago
- ⬢github new repos, most starred 1 d ago
API: https://socialmediatrends-api.osmike.com/v1/repos/N1nEmAn/sapper-v2