requesting-code-review
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute local
gitcommands (git rev-parse,git log,git diff) to determine the commit range and extract code changes for review. - [INDIRECT_PROMPT_INJECTION]: As a code review tool, the skill is naturally exposed to untrusted data within the code changes it processes.
- Ingestion points: The
code-reviewer.mdtemplate instructs the subagent to read the output ofgit diffbetween two commit SHAs. - Boundary markers: None; the code changes are processed without specific delimiters or instructions to ignore embedded prompts in comments or strings.
- Capability inventory: The skill uses the
Tasktool and shell access to perform git operations and review file contents. - Sanitization: None; the subagent operates directly on the raw code diff provided by the git CLI.
Audit Metadata