The Agent Skills Directory

[PROMPT_INJECTION]: The skill has an indirect prompt injection surface. It retrieves untrusted data from Webflow sites and can execute potentially destructive actions.
Ingestion points: Data is ingested through WEBFLOW_GET_PAGE_DOM, WEBFLOW_LIST_COLLECTION_ITEMS, and WEBFLOW_LIST_ORDERS.
Boundary markers: The skill lacks instructions to delimit external content or treat it as untrusted.
Capability inventory: Significant capabilities include WEBFLOW_CREATE_COLLECTION_ITEM, WEBFLOW_DELETE_COLLECTION_ITEM, and WEBFLOW_PUBLISH_SITE.
Sanitization: No evidence of data sanitization or validation before processing was found.
[EXTERNAL_DOWNLOADS]: The skill directs the user to add an external MCP server endpoint https://rube.app/mcp for its functionality.

webflow-automation