revision
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided manuscripts, creating a surface for untrusted data. The analysis is strictly diagnostic, focused on word counts, scene structure, and checklists. The skill lack network access or arbitrary command execution capabilities, preventing any malicious payload within a manuscript from affecting the system or exfiltrating data.
- [COMMAND_EXECUTION]: The skill includes a local utility script,
scripts/revision-audit.ts, designed to be executed via Deno. The script performs benign text analysis and word counting. It is restricted by Deno's security model, requiring explicit--allow-readpermissions for file access, and does not perform any unauthorized system operations or network requests.
Audit Metadata