cy-review-round

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to ingest and process untrusted data from the repository during the review process.\n
  • Ingestion points: The skill reads source code identified through git diff, requirements documentation in _prd.md and _techspec.md, task definitions in _tasks.md, and architectural decision records in .compozy/tasks/<name>/adrs/.\n
  • Boundary markers: There are no instructions to use delimiters or specific ignore patterns to distinguish between the content being reviewed and the agent's operational instructions.\n
  • Capability inventory: The skill performs shell command execution (git diff, make lint, cy-final-verify) and local filesystem write operations for report generation.\n
  • Sanitization: No mechanisms for validating or escaping the content read from external files are specified before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 11:08 AM
Security Audit — agent-trust-hub — cy-review-round