Skills
skills/nexu-io/open-design/editorial-collage-deck/Gen Agent Trust Hub

editorial-collage-deck

Pass

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection (specifically cross-site scripting within the generated artifact) due to the direct interpolation of untrusted content into HTML.\n
  • Ingestion points: User-controlled text from inputs.json (such as title, body, and bullets) is ingested by the scripts/compose.ts generator script.\n
  • Boundary markers: Absent; the template does not use delimiters or include instructions for the agent to ignore potentially malicious embedded content in the input fields.\n
  • Capability inventory: The skill utilizes file-write to generate the HTML file and node-runtime to execute the composition logic.\n
  • Sanitization: Absent; the script interpolates input strings directly into the HTML structure without escaping or validating the content, allowing for the inclusion of arbitrary HTML tags.\n- [EXTERNAL_DOWNLOADS]: The skill's output references external resources from well-known services.\n
  • The generated HTML files link to Google Fonts services (fonts.googleapis.com and fonts.gstatic.com).
Audit Metadata
Risk Level
SAFE
Analyzed
May 4, 2026, 05:45 AM
Security Audit — agent-trust-hub — editorial-collage-deck