executing-plans
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external plan files which could contain malicious instructions. Ingestion points: The agent reads plan files from the local workspace as part of Step 1 in SKILL.md. Boundary markers: The instructions in SKILL.md require the agent to review the plan critically and raise concerns before execution, though no technical delimiters are enforced. Capability inventory: The skill is authorized to perform implementation tasks, including code modification and verification steps. Sanitization: The process relies on the agent's critical analysis and human approval rather than automated sanitization of the input file.
Audit Metadata