requesting-code-review
Warn
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The template in
code-reviewer.mduses string interpolation to build shell commands:git diff --stat {BASE_SHA}..{HEAD_SHA}andgit diff {BASE_SHA}..{HEAD_SHA}. This pattern is vulnerable to command injection if the SHA variables contain shell metacharacters or are sourced from untrusted input. - [PROMPT_INJECTION]: The skill processes untrusted input through placeholders like
{DESCRIPTION}and{PLAN_REFERENCE}incode-reviewer.mdwithout safety markers. - Ingestion points: Input variables in
code-reviewer.mdfrom external implementation descriptions and plans. - Boundary markers: No delimiters or "ignore instructions" warnings are present to distinguish system instructions from external data.
- Capability inventory: Execution of shell-based git commands and making critical decisions on code merge readiness.
- Sanitization: No input validation or escaping is applied to the data before it is inserted into the prompt.
Audit Metadata