skills/ghaida/intent/blueprint/Gen Agent Trust Hub

blueprint

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues were detected. The skill follows expected practices for a design and systems architecture tool. It uses Model Context Protocol (MCP) to interact with well-known, legitimate services (Figma and Pencil.dev) and generates self-contained HTML for local visualization without external dependencies.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data (e.g., support tickets, architecture reviews) to create system models, which technically presents an indirect prompt injection surface.
  • Ingestion points: User-provided service evidence and technical documentation processed during mapping tasks.
  • Boundary markers: The skill does not explicitly define delimiters to isolate external content from its internal instructions.
  • Capability inventory: The skill can generate self-contained HTML code blocks and call external design tools via mcp__claude_ai_Figma__use_figma and mcp__pencil__batch_design.
  • Sanitization: No explicit sanitization or validation of external input content is described in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:34 AM
Security Audit — agent-trust-hub — blueprint