requirements

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection. * Ingestion points: The skill reads project context from an external file named docs/vision.md. * Boundary markers: There are no instructions or delimiters provided to the agent to prevent it from executing commands that might be embedded in the vision document. * Capability inventory: The skill allows the agent to write to the file system (docs/requirements.md) and use the TodoWrite tool based on processed input. * Sanitization: The skill lacks any mechanism to validate or sanitize the content ingested from the vision document before it is used to generate new documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 03:48 PM