executing-plans
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and execute instructions from a 'plan file', which represents an ingestion point for untrusted data. While this is the intended functionality of the skill, it creates an attack surface where a maliciously crafted plan could influence the agent's behavior.
- Ingestion points: Processes tasks from a plan file in Step 1 and Step 2.
- Boundary markers: None explicitly mentioned for isolating plan content.
- Capability inventory: File writes (TodoWrite), task execution, and repository management (via git-worktrees integration).
- Sanitization: The skill relies on a 'critical review' by the agent before execution to mitigate risks.
Audit Metadata