gc-review-im

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local shell commands to facilitate its review workflow.
  • Evidence: Uses git rev-parse and git diff to identify changes in the repository.
  • Evidence: Uses cat and jq to read and validate JSON configuration files from .gc-review/config.json and ~/.gc-review/config.json.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it analyzes untrusted content from source code files and git history.
  • Ingestion points: Content of files identified as IM-relevant (SQL, Prisma, Python models, etc.) and git diff outputs.
  • Boundary markers: Absent. The instructions do not provide delimiters or negative constraints to prevent the model from following natural language instructions found within the code comments or data structures being reviewed.
  • Capability inventory: Local command execution (git, cat), access to the local git repository history, and generation of compliance reports.
  • Sanitization: Absent. The skill processes raw strings from code changes directly through its analysis logic without escaping or filtering potential injection patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 06:23 AM