plan-writing

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • Indirect Prompt Injection (LOW): Ingestion point: Reads project files via Read, Glob, and Grep tools; Boundary markers: Absent; Capability inventory: Read-only file system access; Sanitization: Absent. The skill acts as a formatter for external data with no side-effect capabilities.
  • Tool Access (SAFE): The skill only allows Read, Glob, and Grep tools, which are inherently limited to non-modifying operations.
  • Command Execution (SAFE): Shell command examples in the documentation (e.g., npx, curl) are provided as instructional text for agent planning and are not executed by the skill itself.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 06:26 AM