webflow-hello-world

Pass

Audited by Gen Agent Trust Hub on Mar 25, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external data from Webflow sites and collections.\n
  • Ingestion points: The code snippets list sites, collections, and items from the Webflow API (webflow.sites.list, webflow.collections.list, webflow.collections.items.listItems), bringing external content into the agent's context.\n
  • Boundary markers: Data retrieved from the API is printed directly to the console without delimiters or instructions for the agent to disregard any embedded commands.\n
  • Capability inventory: The skill is configured with Bash, Write, and Edit tools, which could be misused if the agent obeys instructions found within the retrieved content.\n
  • Sanitization: The example scripts do not perform sanitization or validation of the API field data before outputting it.\n- [EXTERNAL_DOWNLOADS]: The skill utilizes the webflow-api package, which is the official SDK hosted on the standard NPM registry.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 25, 2026, 04:42 PM