web-development

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references multiple markdown files and skill manifests hosted on cnb.cool (e.g., Change Safety Protocol, Deployment Gate). These resources are located within the tencent/cloud/cloudbase repository, which is consistent with the skill's author and purpose.
  • [COMMAND_EXECUTION]: The instructions direct the agent to execute standard development commands such as npm run build, vite build, tsc --noEmit, and eslint to verify code quality and build integrity. These are necessary steps for the stated engineering goals.
  • [INDIRECT_PROMPT_INJECTION]: The skill incorporates the use of the agent-browser tool to visit and interact with web pages for validation. This introduces a surface where the agent processes external, untrusted content from the browser, though this is an inherent and documented part of the skill's validation workflow.
  • [SAFE]: All external SDK references, such as @cloudbase/js-sdk, are official vendor packages. The skill explicitly discourages the use of 'any' in TypeScript to maintain code safety and requires manual verification steps before completion.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 04:24 PM
Security Audit — agent-trust-hub — web-development