makeui

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data by ingesting API responses and metadata to generate UI components. \n
  • Ingestion points: API calls in src/lib/service-api (such as /api/users and /api/departments) and host-provided object/field metadata. \n
  • Boundary markers: Mandatory use of UI states (loading, empty, error, forbidden) and field-type display adapters to wrap and normalize content. \n
  • Capability inventory: The skill guides the generation of React components that perform data rendering, routing, and API interaction. \n
  • Sanitization: Instructions mandate the use of normalized field display adapters, numeric validation (decimal-place limits), and type-appropriate controls to prevent raw data exposure and malformed inputs. \n- [COMMAND_EXECUTION]: The skill includes Node.js test scripts (scripts/test-*.mjs) designed to verify the consistency of the skill's own instructions and documentation contracts. These scripts only perform local file system read operations within the skill directory. \n- [EXTERNAL_DOWNLOADS]: The instructions reference several external packages under the @qfei-design/ scope, which are recognized as vendor-owned resources for the authoring entity and are consistent with the intended UI functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 07:07 AM
Security Audit — agent-trust-hub — makeui