webflow-automation
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from Webflow sites, which could potentially contain malicious instructions for the agent.
- Ingestion points: Untrusted data enters the agent context through tools like WEBFLOW_LIST_COLLECTION_ITEMS, WEBFLOW_GET_PAGE_DOM, and WEBFLOW_LIST_ORDERS (SKILL.md).
- Boundary markers: The skill instructions do not define delimiters or provide 'ignore embedded instructions' warnings for data retrieved from external sources.
- Capability inventory: The skill provides write access to the Webflow environment via tools like WEBFLOW_CREATE_COLLECTION_ITEM, WEBFLOW_UPDATE_COLLECTION_ITEM, WEBFLOW_PUBLISH_SITE, and WEBFLOW_DELETE_COLLECTION_ITEM.
- Sanitization: No mechanisms for sanitizing, escaping, or validating external content are described.
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to configure a remote Model Context Protocol (MCP) server.
- The setup section requires adding the external endpoint https://rube.app/mcp to the client configuration to enable the toolkit functionality.
Audit Metadata