full-stack-panel-authoring

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes standard shell utilities including find, grep, and cat to facilitate project structure discovery and architectural pattern identification.
  • [DATA_EXPOSURE]: Service layer templates include standard implementation patterns for retrieving authentication tokens from localStorage to populate API request headers.
  • [PROMPT_INJECTION]: The skill establishes an ingestion surface for indirect prompt injection by reading local project files to detect server dispatch mechanisms and coding styles. Ingestion points: Project source files (e.g., server/index.ts) read via cat. Boundary markers: Absent. Capability inventory: File system discovery, reading, and writing. Sanitization: Not applicable as the analysis is structural rather than execution-based.
  • [COMMAND_EXECUTION]: The Panel base class uses innerHTML for UI rendering; however, the skill explicitly warns the developer against utilizing this method with untrusted data in the implementation logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 11:24 AM
Security Audit — agent-trust-hub — full-stack-panel-authoring