fusion-design

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructs the agent to treat a local file (DESIGN.md) as the authoritative source for design decisions, overriding standard rules.
  • Ingestion points: DESIGN.md located at the root of the application project being modified.
  • Boundary markers: None provided. The instructions explicitly state that the external file is the authoritative source and overrides or supplements other reference rules.
  • Capability inventory: The skill is designed for frontend UI implementation and code generation. No scripts or tools for network access or arbitrary command execution were found in the skill content.
  • Sanitization: No sanitization or validation of the content within the DESIGN.md file is described or performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 09:00 AM
Security Audit — agent-trust-hub — fusion-design