Skills
Navigate
skills/broomva/skills/broomva-design/Gen Agent Trust Hub

broomva-design

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The SKILL.md file instructs agents to execute a local Python script scripts/materialize.py to materialize design profiles. This is the primary function of the skill.
  • [EXTERNAL_DOWNLOADS]: Multiple HTML specimen files in assets/system (e.g., forms.card.html, navigation.card.html) download the React, React-DOM, and Babel libraries from unpkg.com, which is a well-known public package CDN.
  • [PROMPT_INJECTION]: The assets/system/apps/maestro/WorkDetail.jsx component uses the dangerouslySetInnerHTML property to render the html field of assistant messages, creating a surface for indirect prompt injection. 1. Ingestion points: Data entering through the item.chat structure. 2. Boundary markers: None present for the HTML content. 3. Capability inventory: Ability to render arbitrary HTML. 4. Sanitization: No explicit sanitization or escaping is implemented in the component code.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 01:15 AM
Security Audit — agent-trust-hub — broomva-design