start-dev
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses grep to find task entries. User-provided task IDs are normalized and validated against known prefixes, which helps prevent command injection.
- [DATA_EXPOSURE]: Reads docs/MASTER_PLAN.md to manage task statuses, aligning with its intended purpose.
- [PROMPT_INJECTION]: Evaluated for indirect injection risk since it processes external markdown files. 1. Ingestion points: Reads task status and context from docs/MASTER_PLAN.md. 2. Boundary markers: None present. 3. Capability inventory: File search and modification. 4. Sanitization: Normalizes task IDs to uppercase and checks against allowed prefixes.
Audit Metadata