coding-guidance-python
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes Python source code and metadata, which is a potential surface for indirect injection. However, the instructions include specific safety requirements such as input validation and safe subprocess argument handling.
- Ingestion points: The Implementation Workflow in SKILL.md requires reading code modules and package metadata.
- Boundary markers: The language rules specifically advise validating external input at the boundary.
- Capability inventory: Instructions cover file system access, network I/O, and subprocess execution.
- Sanitization: Service boundary rules mandate the use of separate argument elements for subprocesses to prevent command injection.
Audit Metadata