terminology-normalizer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [Prompt Injection] (SAFE): The skill uses instructional role-play (Taxonomist, Integrator) to define behavior. No malicious override or safety bypass patterns were detected.
- [Data Exposure & Exfiltration] (SAFE): The skill operates on local project files and declares no network access. No interaction with sensitive system files or credentials was found.
- [Indirect Prompt Injection] (LOW): The skill reads and modifies
output/DRAFT.md, creating a surface for indirect prompt injection if the draft contains malicious instructions. However, the risk is minimized by strict output constraints. Ingestion points:output/DRAFT.md,outline/outline.yml,outline/taxonomy.yml. Boundary markers: Absent. Capability inventory: File-write (output/DRAFT.md). Sanitization: Absent. - [Remote Code Execution] (SAFE): No remote scripts, external dependencies, or dynamic code execution methods are used.
Audit Metadata