documentation-writer
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [Prompt Injection] (SAFE): The skill contains instructional prompts for documentation formatting and style but does not include any bypass, override, or system prompt extraction commands.
- [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, hardcoded credentials, or network operations were detected. Interactions are limited to the project's own documentation.
- [Remote Code Execution] (SAFE): The skill consists entirely of markdown instructions and does not reference external packages, scripts, or dynamic code execution methods.
- [Indirect Prompt Injection] (LOW): The skill identifies a surface where untrusted data could influence agent behavior. 1. Ingestion points: The skill triggers on file edits and new pages within the codebase. 2. Boundary markers: No explicit markers or 'ignore' instructions are present to delimit the code being analyzed. 3. Capability inventory: The agent has the capability to write to SITE.md. 4. Sanitization: There is no evidence of sanitization for comments or strings extracted from the code. Risk: Malicious comments in a code change could attempt to manipulate the summary in the documentation, but the impact is restricted to text content in SITE.md.
Audit Metadata