code-review
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands (
git diff --staged,git diff) to retrieve source code changes for analysis. These are standard read operations for a code review tool. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing untrusted data (code diffs and patches) that could contain malicious instructions meant to subvert the agent's behavior.
- Ingestion points: Data is ingested from
git diffcommand outputs and user-provided patch/diff strings in theSKILL.mdfile. - Boundary markers: Absent. The instructions do not define clear delimiters or provide the agent with "ignore embedded instructions" warnings for the content being reviewed.
- Capability inventory: The skill uses
gitto read repository states. - Sanitization: Absent. There is no evidence of input validation, escaping, or filtering of the code content before it is processed by the AI.
Audit Metadata