converge
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill contains no executable scripts, hardcoded credentials, or network operations. It functions entirely through natural language instructions guiding the agent's design behavior.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user preferences and external references to guide design changes, which technically constitutes an indirect prompt injection surface. However, this is the core intended functionality for a design refinement tool and is mitigated by instructions requiring the agent to be surgical and ask for clarification rather than guessing intent.
- Ingestion points: User preferences and shared references (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: File reading and local file writing (implied by Rule 6 and design editing instructions).
- Sanitization: Absent.
Audit Metadata