executing-plans
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and execute instructions from an external source such as plan.md files.
- Ingestion points: The skill reads external plan files (e.g., docs/plan.md) in Step 1 to define the agent's tasks.
- Boundary markers: The instructions lack technical boundary markers or sandboxing for the loaded content, relying instead on procedural 'critical review' steps by the AI.
- Capability inventory: The skill grants the agent capabilities to modify the file system, execute shell commands for testing (e.g., npm test, pytest), and manage git repositories.
- Sanitization: There is no automated validation or sanitization of the plan content before the agent interprets it as executable tasks.
Audit Metadata