code-review-checklist
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze untrusted data from external sources such as pull request descriptions, linked issue trackers, and modified source code files. \n- Ingestion points: Steps 1 and 2 in
SKILL.mddirect the agent to read PR descriptions, tickets, and file changes. \n- Boundary markers: The instructions do not define clear delimiters or provide "ignore instructions" directives when processing these untrusted inputs. \n- Capability inventory: The skill leverages the agent's core capabilities for file reading and context analysis on attacker-controllable data. \n- Sanitization: There are no provisions for sanitizing or escaping the content being reviewed before it is processed by the model.
Audit Metadata