close
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill outlines Git operations (stage, commit, push) and file modifications for project state management. These are scoped to authorized contexts and verified work artifacts.- [DATA_EXFILTRATION]: Workflow state and session handoffs are stored locally in the ".lev/pm/" directory. No network exfiltration or unauthorized file access patterns were detected.- [INDIRECT_PROMPT_INJECTION]: The skill processes work templates and session files. It addresses potential surface risks through a "recall check" validation step to ensure consistency between handoff summaries and original source artifacts.
Audit Metadata