stitch-ui-variants
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust security-oriented workflow (Pre-flight, Planned, Executing, Verifying) and explicitly requires user authorization for any irreversible actions.
- [SAFE]: No obfuscation, hardcoded credentials, or dangerous remote code patterns were found in the skill instructions or configuration files.
- [SAFE]: The skill identifies itself as a 'Logic skill' and does not request or demonstrate unnecessary capabilities like network access or persistent file system modifications.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided design specs ('Base Spec') to generate prompts. While this is a potential ingestion point for indirect instructions, the skill includes comprehensive safety rules (Least Privilege, Privacy Security) and a multi-step validation workflow that mitigates this risk. Severity is low as this is an inherent property of text processing.
- Ingestion points: 'Base Spec' input processed in SKILL.md.
- Boundary markers: Explicit validation checklist and pre-flight check requirements are included in the workflow.
- Capability inventory: No subprocess calls, file-writes, or network operations are present in the skill files.
- Sanitization: Instructions explicitly mandate privacy security and credential protection in logs and outputs.
Audit Metadata