// Software stack
Security & red-team agents
Authorized vuln discovery, skill audits and agent defenses.
4 pieces4 setup stepsscenario: Let agents run safely
What goes in it
- 1 ★ 21k
SkillSpector Identity, secrets & agent security Open source
Scanning
Security scanner for AI agent skills.
Or: agent-scan · AI-Infra-Guard · deepsec
- 2 ★ 49k
shannon Identity, secrets & agent security Open source
Pen-testing (authorised)
Autonomous white-box AI pentester for web apps and APIs.
Or: hexstrike-ai · Decepticon
- 3 ★ 7.7k
trailofbits/skills Skills & rules Open source
Security skills
Trail of Bits Claude Code skills for security research.
- 4 ★ 6.4k
agent-governance-toolkit Identity, secrets & agent security Open source
Governance
AI Agent Governance Toolkit — policy, zero-trust identity, sandboxing.
How to set it up
- 01Get the scope authorised in writing
- 02Scan skills and MCP servers first
- 03Run pen-test agents on an isolated network
- 04Share reports only with the security lead
⚠︎ For authorised testing only; unauthorised attacks are illegal.