review-ready-code
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on shell-based Git commands to perform its core functions. It uses commands such as
git rev-parse,git diff, andgit range-diffto capture tree states, verify identity invariants, and manage branch projections. - [EXTERNAL_DOWNLOADS]: At every invocation, the skill fetches content from Google's official Engineering Practices documentation on GitHub (
google.github.io). This documentation is used to dynamically inform the agent's strategy for commit decomposition. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface where instructions could be hidden in processed data.
- Ingestion points: Local repository diffs (code hunks) and external guidance fetched from the Google Engineering Practices URL.
- Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore' instructions when processing the diff content or the fetched HTML/Markdown.
- Capability inventory: The agent has the ability to execute Git commands, create/delete branches, and manage worktrees within the local filesystem.
- Sanitization: Absent. The skill does not describe any methods for filtering or escaping instructions that might be embedded within the code diffs it analyzes.
Audit Metadata