webflow-mcp:asset-audit
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external data from Webflow sites, including filenames, alt text, and image content (via
get_image_preview). This creates a surface for indirect prompt injection where instructions embedded in asset metadata or filenames could potentially influence agent behavior. - Ingestion points: Webflow site asset data and image previews fetched from the target site's environment.
- Boundary markers: The instructions do not specify delimiters or warnings to the agent to ignore instructions that might be contained within the analyzed asset content.
- Capability inventory: The skill possesses the capability to modify site assets using the
asset_toolbased on the results of the data analysis. - Sanitization: The skill implements formatting rules for output (e.g., SEO-friendly filenames), but does not explicitly filter or sanitize input data for malicious prompt injection patterns.
Audit Metadata