lov-code-review
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of code diffs and project files, which creates a potential surface for indirect prompt injection attacks.
- Ingestion points: The skill reads file content and git diffs from the target repository as specified in
SKILL.md(Workflow Step 1). - Boundary markers: There are no explicit instructions in
SKILL.mdor the scripts to wrap analyzed content in delimiters or ignore embedded instructions within the code comments being reviewed. - Capability inventory: The skill utilizes host-provided capabilities including file reading, searching, CLI execution, and browser access as described in the 'Execution boundary' section of
SKILL.md. - Sanitization: The skill lacks explicit sanitization or filtering of instructions that might be embedded within the source code being analyzed.
Audit Metadata