weaverse-integration

Pass

Audited by Gen Agent Trust Hub on Nov 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to analyze local project files and fetch external documentation, creating a surface for potential injection of malicious instructions.
  • Ingestion points: Project configuration files (package.json, server.ts, root.tsx) and dynamic search results from documentation scripts.
  • Boundary markers: The skill does not provide delimiters or instructions to ignore instructions embedded within analyzed files.
  • Capability inventory: The agent performs file reading, shell command execution (grep, cat, node, npm), and filesystem modifications.
  • Sanitization: No logic for sanitizing or validating external content is specified.
  • [COMMAND_EXECUTION]: The skill uses standard development tools like grep and cat to inspect the project environment and executes Node.js scripts to retrieve documentation.
  • [EXTERNAL_DOWNLOADS]: The skill installs the official @weaverse/hydrogen package and related helper tools from the vendor's official repositories.
  • [DYNAMIC_EXECUTION]: The skill's recommended Content Security Policy (CSP) includes 'unsafe-eval' specifically for use in development/design mode, which is used by the vendor's Studio bridge for dynamic schema processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 15, 2026, 06:57 PM