repo-security-posture

Audits GitHub Actions workflows for supply-chain and CI/CD security vulnerabilities including dangerous triggers, template injection, mutable action pins, cache poisoning, and release hardening, with severity ratings and concrete rewrites.

Evaluates whether an AI agent skill, plugin, or tool is safe to install by running a two-stage analysis: a deterministic offline scanner (regex, AST, taint tracking, YARA) followed by an LLM intent and contract verification check.

Skill

Skill

A security skill for AI coding agents that provides structured workflows for supply chain security auditing, distilled from real security work on codebases serving 100M+ users per week. It loads automatically when the agent hits a relevant task.

Security skills for AI coding agents distilled from real-world security work on codebases serving 100M+ users, covering offensive engagement planning, LLM red-team testing with training data generation, and CI/CD supply-chain security auditing.

Security skills for AI coding agents distilled from real security work on codebases serving 100M+ weekly users. Includes offensive engagement planning, red-team workflows with phase gates, LLM guardrail training data generation, and exploitability validation, all using the open Agent Skills format.

Skill

Skill

Skill