codex-review
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to process external source code and commit history for reviews and changelog generation, which creates an indirect prompt injection surface.
- Ingestion points: The skill ingests untrusted project code and commit messages during the review process (SKILL.md).
- Boundary markers: Absent. No delimiters or instructions are provided to the agent to distinguish between code and potential malicious instructions embedded within that code.
- Capability inventory: The skill analyzes source code and generates text content for project documentation (SKILL.md).
- Sanitization: Absent. There is no mention of filtering or sanitizing the external content before processing.
- [EXTERNAL_DOWNLOADS]: The installation guide references an external third-party repository (
BenedictKing/codex-review) which is not associated with a trusted vendor or the designated author. - [NO_CODE]: The skill consists entirely of markdown instructions and does not include any local scripts, binaries, or executable code, reducing the direct execution risk from the skill package itself.
Audit Metadata