writing-plans
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security threats such as exfiltration, privilege escalation, or unauthorized command execution were found. The skill follows legitimate software engineering practices.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided specification documents to generate and review implementation plans. This is a functional requirement of the skill but introduces a potential attack surface.
- Ingestion points: Specification and requirement files identified in the planning and review instructions.
- Boundary markers: The skill lacks explicit boundary markers or instructions to disregard embedded commands in the input data.
- Capability inventory: The resulting plans include file system modifications and execution of standard development tools like git and pytest.
- Sanitization: No explicit sanitization of input text is performed, relying on model-level controls.
Audit Metadata