quality-assurance-auditor
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local Python script to manage project directories, generate a task manifest, and check the completion status of draft files. These operations are confined to the local project workspace.
- [PROMPT_INJECTION]: The instructions employ strong directive language to enforce gatekeeping behavior, requiring the agent to block further progress if audits fail. This is a functional requirement of the skill's role and does not represent an attempt to subvert AI safety filters.
- [DATA_EXFILTRATION]: The skill reads project-related files to perform auditing functions but lacks network capabilities or commands that would allow this data to be transmitted externally.
- [EXTERNAL_DOWNLOADS]: Mentions of tools like pandoc or python-docx for file conversion are standard for document processing workflows and do not involve suspicious remote resources.
Audit Metadata