graphql-api-design

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely informational and contains no malicious instructions or suspicious code patterns.
  • [COMMAND_EXECUTION]: Code snippets demonstrate GraphQL resolver logic and batch loading. They do not involve subprocess spawning or unsafe shell execution.
  • [PROMPT_INJECTION]: The content is focused on technical documentation and does not attempt to manipulate the AI agent's core instructions or safety guardrails.
  • [EXTERNAL_DOWNLOADS]: No remote URLs for script execution or binary downloads were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 04:55 AM