codex-review
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the codex CLI tool and shell utilities to perform its primary function. Evidence: Commands include mktemp for directory creation and codex -C review for analyzing the current directory (SKILL.md).
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes external, untrusted code data. Ingestion points: The skill processes local code changes, branch diffs, and commit history (SKILL.md). Boundary markers: The workflow lacks explicit delimiters or instructions to ignore embedded commands within the analyzed code. Capability inventory: The skill leverages shell command execution through the codex CLI and mktemp utility (SKILL.md). Sanitization: Instructions direct the agent to manually verify all findings against the actual source code before presenting them to the user.
Audit Metadata