webflow-automation

Warn

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires adding an external Model Context Protocol (MCP) server at https://rube.app/mcp to provide the tools necessary for the Webflow automation workflows.
  • [COMMAND_EXECUTION]: The skill enables the agent to perform management tasks on Webflow sites, such as WEBFLOW_PUBLISH_SITE, WEBFLOW_CREATE_COLLECTION_ITEM, and WEBFLOW_DELETE_COLLECTION_ITEM, all executed via the remote Rube MCP server.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection. * Ingestion points: Data is ingested from external Webflow sites through tools like WEBFLOW_LIST_COLLECTION_ITEMS, WEBFLOW_GET_PAGE, and WEBFLOW_GET_PAGE_DOM (SKILL.md). * Boundary markers: No specific boundary markers, delimiters, or instructions are provided to help the agent distinguish between legitimate site data and potentially malicious instructions embedded within that data. * Capability inventory: The skill possesses significant capabilities, including the ability to create, modify, and permanently delete CMS items and publish live site changes (SKILL.md). * Sanitization: The instructions do not define any sanitization or validation processes for the data retrieved from Webflow before it is used in subsequent agent operations.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 18, 2026, 03:49 PM