safe-publish
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust security model for administrative actions by requiring a specific confirmation keyword ('publish') instead of generic affirmative responses, which prevents accidental site deployment.
- [SAFE]: It follows the principle of least privilege by explicitly restricting the agent to use only designated Webflow MCP tools and prohibiting unauthorized methods.
- [SAFE]: The workflow includes comprehensive pre-publish validation and post-publish verification steps to ensure site integrity and accessibility.
- [SAFE]: All resource references, including domains (webflow.io) and tool names, are consistent with the official infrastructure of the vendor (Webflow).
Audit Metadata