devtools
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- Indirect Prompt Injection Surface: The tool processes AI chat messages to extract and render UI specifications. This creates a surface where external data could potentially influence the agent's behavior. 1. Ingestion points: The messages property in SKILL.md. 2. Boundary markers: None identified in instructions. 3. Capability inventory: State modification via store.set. 4. Sanitization: Not explicitly documented.
- Command Execution: The skill provides instructions for installing dependencies using standard package managers.
- Runtime State Modification: The inspector allows for editing application state directly. This is an intended feature for debugging that provides significant control over the application's runtime data.
- Security Isolation: The tool employs Shadow-DOM isolation to prevent styles from leaking between the panel and the host application, and it includes logic to disable rendering in production environments.
Audit Metadata