full-code-review
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data (git diffs and PR metadata) and interpolates it directly into the prompts for its reviewer agents.
- Ingestion points: The
DIFFandCHANGED_FILESvariables inscripts/full-code-review.jsare populated with external repository content. - Boundary markers: The diff content is wrapped in markdown code blocks ("```diff"), which provide some structural separation but do not prevent an LLM from following instructions embedded within the diff.
- Capability inventory: The skill orchestrates multiple LLM calls (
agent()) and utilizes read-only shell commands viagitandgh. - Sanitization: No explicit sanitization or instruction-filtering is applied to the diff content before it is passed to the reviewer models.
Audit Metadata