executing-plans
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions do not contain any obfuscated commands, remote code downloads, or unauthorized data access patterns. It primarily functions as a workflow guide for task management.- [PROMPT_INJECTION]: The skill ingests external data from plan files, which constitutes an indirect prompt injection surface. This risk is minimized by the mandatory critical review step and instructions to stop execution immediately if instructions are unclear. Ingestion points: External plan files (SKILL.md, Step 1). Boundary markers: None defined. Capability inventory: General agent file-writing and command execution capabilities. Sanitization: Procedural review and human-partner checkpoints.
Audit Metadata