security-engineer

Prompt injection specialist — adversarial testing of LLM-powered applications for prompt injection, jailbreaks, data extraction, and indirect injection. Use when security-testing AI integrations, evaluating guardrail robustness, or assessing LLM attack surface in production systems.

Security engineer — threat modelling, security audits, compliance, vulnerability management. Use for security reviews, threat models, compliance documentation, or dependency vulnerability triage.

Bootstrap the security documentation structure for a project. Creates docs/security/, generates initial templates and root SECURITY.md, and writes domain CLAUDE.md. Idempotent — merges missing sections into existing files without overwriting.

Audit project dependencies for known vulnerabilities, outdated packages, and license issues.

Passive reconnaissance on a target domain or organisation using open-source intelligence. Maps the attack surface from publicly available sources only. Use at the start of a penetration test or security assessment to understand what's exposed before active testing begins.

Review code or configuration for security vulnerabilities — OWASP Top 10, secrets, auth, injection.

Audit the software supply chain for integrity risks — source, build, dependencies, and artifact provenance. Produces a SLSA-aligned assessment with findings and hardening recommendations. Use when assessing supply chain posture or after a dependency incident.

1 hook across 1 event