update-claude-documentation
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill focuses on documentation management within the local repository. It performs standard file-read and file-write operations on markdown files to ensure consistency and clarity.\n- [SAFE]: No network activity or external data exfiltration. The skill does not reference external URLs or attempt to transmit data outside the project environment.\n- [SAFE]: No code execution risks. The skill does not involve dynamic execution, package management, or script downloads.\n- [PROMPT_INJECTION]: The documentation update process presents an indirect prompt injection surface as it incorporates project data and session history into files. This risk is classified as safe as the skill only generates documentation and lacks high-risk capabilities.\n
- Ingestion points: Existing repository documentation and session context.\n
- Boundary markers: Not explicitly defined in the templates.\n
- Capability inventory: File-write access to markdown files in the repository.\n
- Sanitization: No specific filtering or validation logic is applied to the data being documented.
Audit Metadata