project-profile

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting data from external sources to influence project configuration.
  • Ingestion points: Processes configuration from a user-global file (~/.aidoc/profile.yaml) and accepts input via an interactive user interview.
  • Boundary markers: The skill lacks explicit boundary markers or directives to ignore embedded instructions when writing the resulting .aidoc/profile.yaml file.
  • Capability inventory: The skill is authorized to write to the local file system to maintain project profiles.
  • Sanitization: The skill performs schema validation against a framework registry (ADAPTATION_SURFACE.yaml), filtering out keys not explicitly defined in the allowed surface.
  • [NO_CODE]: The skill consists only of instructional content and markdown-based logic without accompanying executable scripts or binaries.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 08:56 PM
Security Audit — agent-trust-hub — project-profile