skills/howells/agentify/agents/Gen Agent Trust Hub

agents

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to check for installed Node.js packages during the project detection phase. This is a legitimate use case for a scaffolding tool to determine the current project environment.
  • [EXTERNAL_DOWNLOADS]: The skill suggests the installation of well-known and framework-specific Node.js packages such as @mastra/core, zod, and various AI SDK providers. These are industry-standard dependencies for the stated purpose.
  • [CREDENTIALS_UNSAFE]: The skill provides templates for environment-based model routing. It correctly references environment variable names (e.g., GOOGLE_GENERATIVE_AI_API_KEY) rather than hardcoding sensitive credentials. It explicitly warns against passing user IDs or auth tokens directly to tools, recommending a server-side request context instead.
  • [DATA_EXFILTRATION]: While the skill reads project metadata (package.json) and source file structures, it does so to provide context-aware scaffolding. There is no evidence of these findings being sent to external or unauthorized servers.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 02:49 PM