The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied source code for security auditing. Ingestion points: User-provided source code and application logic. Boundary markers: None explicitly defined. Capability inventory: Local file writing to ~/.claude/skill-analytics/ and network access to Have I Been Pwned API. Sanitization: Recommends strict schema validation and HTML sanitization for applications.\n- [EXTERNAL_DOWNLOADS]: Fetches data from the well-known api.pwnedpasswords.com service to verify credential safety against known breaches using privacy-preserving k-anonymity.\n- [COMMAND_EXECUTION]: Recommends the use of industry-standard security auditing tools such as npm audit, pip-audit, and snyk test for dependency scanning.

security-skill