acc-extract-domain-concepts

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION] (INFO): Employs standard filesystem search patterns (grep and glob) to identify DDD components like Entities and Value Objects. These operations are limited to local discovery.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill processes untrusted external content in the form of source code. Evidence Chain: (1) Ingestion: reads PHP files via grep; (2) Boundary markers: none; (3) Capabilities: restricted to analytical documentation (no network/write); (4) Sanitization: none. Severity is low as the output only influences agent reasoning and documentation.
  • [DATA_EXPOSURE] (INFO): Reads code structure to build a business glossary; no sensitive file paths (e.g., .env, credentials) or network exfiltration patterns are present.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 08:42 AM