requesting-code-review
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill is susceptible to indirect prompt injection because it processes untrusted content from git diffs and user-defined implementation summaries without adequate boundary markers.
- Ingestion points: The
code-reviewer.mdtemplate (lines 14, 18, 24) ingests{DESCRIPTION},{PLAN_REFERENCE}, and the results ofgit diffat runtime. - Boundary markers: There are no explicit delimiters or instructions to the subagent to ignore embedded commands within the code or descriptions being reviewed.
- Capability inventory: The subagent primarily performs read-only operations on the local git repository and provides textual feedback; no dangerous write or network capabilities are exposed.
- Sanitization: The skill does not sanitize the input strings or the output of the git commands before passing them to the LLM.
Audit Metadata