ai-writing-audit
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- Indirect Prompt Injection (LOW): The skill performs project-wide scans on markdown and source files (README, docs, src), creating an ingestion point for untrusted data. If these files contain malicious instructions, the agent may be influenced to deviate from the audit task.\n
- Ingestion points: Local file system scanning across project directories (README*, docs/, src/, etc.).\n
- Boundary markers: Absent; the prompt does not provide explicit delimiters or instructions to the agent to ignore embedded commands in the scanned text.\n
- Capability inventory: File read access and file rewrite capabilities (Phase 3).\n
- Sanitization: Absent; the skill processes file content without escaping or validation for instructions.
Audit Metadata