skills/dafang/codestable/cs-trick/Gen Agent Trust Hub

cs-trick

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes source code and existing documentation to create new entries, which creates a surface for indirect prompt injection.\n
  • Ingestion points: Project source code files (Phase 2) and existing documentation in the .codestable/compound/ directory (Phase 1.5).\n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded content are defined in the workflow.\n
  • Capability inventory: The skill is capable of writing documentation files to the local project structure and executing a local Python tool for metadata searches.\n
  • Sanitization: Ingested content from the project environment is not explicitly sanitized or validated before being included in drafts.\n- [COMMAND_EXECUTION]: The skill uses a local Python tool (.codestable/tools/search-yaml.py) to perform filtered searches of the documentation library. This command is a functional component of the skill's library management logic and operates within the local project scope.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 03:25 AM
Security Audit — agent-trust-hub — cs-trick