dev5-review

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the project's codebase to perform its analysis. \n
  • Ingestion points: Reads requirements from .llm/$ARGUMENTS/prd.md, design specifications from .llm/$ARGUMENTS/design.md, and code changes via git diff. \n
  • Boundary markers: No explicit delimiters or instructions are used to distinguish untrusted file content from the agent's instructions. \n
  • Capability inventory: The skill can execute shell commands (git) and trigger subsequent automation steps (/dev6-review-fix, /dev7-pr). \n
  • Sanitization: No sanitization or validation of the ingested content is performed prior to processing. \n- [COMMAND_EXECUTION]: The skill executes shell-based git commands to manage the development workflow. \n
  • Evidence: Runs git diff to retrieve implemented changes and git add -A && git commit to persist the reviewed work.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 12:37 AM
Security Audit — agent-trust-hub — dev5-review