authoring-react

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill's primary function is to enforce React and TypeScript best practices. Analysis of the instructions and examples revealed no malicious patterns, unauthorized network activity, or credential exposure.
  • [INDIRECT_PROMPT_INJECTION]: The skill operates on user-provided code during refactoring and review, which is a common vector for indirect injection. However, the instructions mandate that the agent check for local project rules and request user intervention if conflicts occur, which serves as a defensive boundary.
  • Ingestion points: Existing UI code and project configuration files (e.g., CONTRIBUTING.md, .cursor/rules) accessed during tasks.
  • Boundary markers: Explicit "Hard contract" for conflict resolution with existing project documentation, requiring the agent to ask the user for confirmation.
  • Capability inventory: Local filesystem write access for code creation, editing, and refactoring on the project's UI components.
  • Sanitization: Human-in-the-loop verification for stylistic conflicts serves as the primary mitigation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 01:10 PM
Security Audit — agent-trust-hub — authoring-react