neo-code-review

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/git-diff-reviewer.py executes local git commands using subprocess.run to retrieve repository state, diffs, and untracked files. Arguments are passed as a list (e.g., cmd = ["git", "diff"]), which mitigates shell injection risks.\n- [PROMPT_INJECTION]: The skill processes untrusted source code and Git diffs, creating an attack surface for indirect prompt injection.\n
  • Ingestion points: Source code snippets and Git diff/file content returned by the helper script.\n
  • Boundary markers: No explicit delimiters or safety instructions are used when interpolating external code into the prompt.\n
  • Capability inventory: The skill can execute a local helper script and read arbitrary files from the workspace.\n
  • Sanitization: No input validation or filtering is performed on the ingested code.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 01:13 PM
Security Audit — agent-trust-hub — neo-code-review