factory-review
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown for a code review task. It does not perform any sensitive operations, execute shell commands, or download external resources.
- [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data from GitHub (PR text, comments, and file contents). Evidence chain: 1. Ingestion points: Pull request text and repository files. 2. Boundary markers: Explicit instructions to treat external data as untrusted. 3. Capability inventory: Analysis and review of pull requests. 4. Sanitization: The skill mandates that the agent flags missing validation or quoting of untrusted inputs.
Audit Metadata