brain-audit
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is designed for local documentation auditing and credential sanitization. It uses standard system tools like bash and grep to perform its tasks and does not involve any external network communication or remote code execution.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests markdown content from the workspace to perform its hygiene audit. This ingestion is the primary purpose of the skill and is handled via specific procedures for link verification and secret redaction.\n
- Ingestion points: Documentation and memory files located in .agents/context/, docs/, .memory/wiki/, and the project root as specified in the Step 1 discovery phase.\n
- Boundary markers: The instructions do not define specific boundary markers for the audited content.\n
- Capability inventory: Uses bash, grep, glob, view_file, write_to_file, and replace_file_content to analyze and modify files.\n
- Sanitization: The skill includes an explicit requirement to mask any discovered credentials with [REDACTED] tags.
Audit Metadata