webflow-mcp:safe-publish
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes vendor-authorized tools for all Webflow operations, ensuring operations occur within intended service boundaries.
- [COMMAND_EXECUTION]: All publishing actions are protected by a manual confirmation requirement where the user must explicitly type 'publish', which prevents accidental or automated execution of deployment tasks.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it retrieves and summarizes external site content during the preview phase. 1. Ingestion points: Site metadata, page information, and CMS collection items are retrieved from the Webflow API as described in Phase 2. 2. Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when presenting fetched site content to the agent. 3. Capability inventory: The skill includes site publishing capabilities (data_sites_tool) and network verification of the resulting live URLs (WebFetch). 4. Sanitization: No specific sanitization or escaping logic is defined for the content fetched from the API before it is processed by the agent.
Audit Metadata