status
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill executes a local Python script which uses subprocess.run to call git for identifying the repository root. This is a standard operation using a fixed argument list.
- Indirect Prompt Injection: The skill reads local configuration files to generate its status report. 1. Ingestion points: Data is read from .apache-magpie.lock and agents.md. 2. Boundary markers: The skill includes a non-negotiable output contract for verbatim display. 3. Capability inventory: The skill is restricted to read-only collection, delegating mutations to the setup skill. 4. Sanitization: Risk is managed through verbatim reporting and mandatory human-in-the-loop confirmation for delegated commands.
Audit Metadata