workflow-planner
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes various repository files, such as source code, design specifications, and architecture guides, to generate its output. This creates a surface for indirect prompt injection where instructions embedded in project data could influence the agent's behavior. However, the risk is significantly mitigated by the skill's strict role boundaries: it is explicitly restricted to writing only to specific documentation paths (e.g.,
docs/features/<feature-id>/spec.md) and metadata files (feature_list.json,PROGRESS.md). It is strictly prohibited from creating or modifying executable application code, preventing any such injection from directly affecting system security or runtime behavior.
Audit Metadata