code-review-ai-ai-review
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Python
subprocessmodule and GitHub Action workflow steps to execute local command-line tools includingsonar-scanner,semgrep,codeql, andtrufflehogfor automated code analysis. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by ingesting and processing untrusted external data.
- Ingestion points: Untrusted content enters the agent's context through the
$ARGUMENTSrequirement and the pull requestdiffanddescriptionfields in the review script. - Boundary markers: The LLM prompt templates lack robust delimiters or specific instructions to treat the interpolated code diffs as data rather than instructions.
- Capability inventory: The skill possesses significant capabilities, including executing shell commands via
subprocess, accessing sensitive API keys from environment variables, and posting content to the GitHub API. - Sanitization: The implementation does not include sanitization or escaping of the ingested code content before it is interpolated into LLM prompts.
Audit Metadata