code-review
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze untrusted code from external sources (files, diffs, pull requests), creating a surface for indirect prompt injection.
- Ingestion points: User-provided Luau/Roblox code snippets, files, and pull request diffs described in the 'Finding review flow' section of
SKILL.md. - Boundary markers: The instructions lack specific requirements for the agent to use secure delimiters or explicit 'ignore instructions within data' warnings when processing code content.
- Capability inventory: The skill is scoped to analysis and reporting (Blocker/Correctness/Advisory findings) and explicitly hands off code writing/refactoring to another skill, limiting the direct impact of successful injection on the local filesystem.
- Sanitization: There are no instructions for the agent to sanitize or escape user-provided code comments or strings before they are processed by the model's reasoning engine.
Audit Metadata