code-review

Warn

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The instructions command the agent to run npm run lint and npx tsc --noEmit. These trigger the execution of scripts and binaries defined within the repository being reviewed. If the repository contains malicious configuration (e.g., in package.json), it could lead to arbitrary code execution on the user's system.\n- [PROMPT_INJECTION]: The skill processes external, untrusted code changes which creates a surface for indirect prompt injection.\n
  • Ingestion points: Content retrieved via git diff and view_file (Step 2 of the skill instructions).\n
  • Boundary markers: Absent. The skill does not provide markers to distinguish code data from instructions.\n
  • Capability inventory: File system read access (view_file), version control access (git), and shell execution (npm, npx).\n
  • Sanitization: None. The agent is instructed to critically analyze the content directly without escaping or filtering.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 8, 2026, 06:22 AM