refine-requirement

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill's primary functionality is restricted to local text processing and file management. It does not utilize dangerous tools, perform network operations, or request elevated privileges.
  • [PROMPT_INJECTION]: The skill manages document content that acts as an indirect prompt injection surface.
  • Ingestion points: The skill reads file content from .agents/flow/{file} and reads metadata from .agents/state.json.
  • Boundary markers: The instructions do not define specific delimiters to isolate the PRD content from the agent's instructions, potentially allowing embedded text to influence behavior.
  • Capability inventory: Capabilities are limited to reading and writing project-specific files; the skill cannot access the network or execute shell commands.
  • Sanitization: No sanitization or path validation is performed on the ingested content or the file paths derived from the project state.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 12:05 PM
Security Audit — agent-trust-hub — refine-requirement