requesting-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses git commands like
git rev-parse,git log, andgit diffto determine commit ranges and extract code changes for review. These are standard operations for its intended use case. - [INDIRECT_PROMPT_INJECTION]: The skill ingests local code changes to be processed by an LLM subagent, creating a surface where malicious instructions embedded in code comments could attempt to manipulate the review output.
- Ingestion points: Git diff content is ingested into the subagent template in
code-reviewer.md. - Boundary markers: The template uses Markdown headers to separate context and instructions.
- Capability inventory: The skill executes local git CLI commands via shell.
- Sanitization: No specific sanitization or filtering of the diff content is mentioned.
Audit Metadata