design-md-to-shadcn
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches design specifications from external URLs as defined in the collection workflow.
- Evidence: SKILL.md (Step 1) specifies fetching raw markdown from URLs using tools like WebFetch.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted external data to perform file-writing operations.
- Ingestion points: External DESIGN.md files are retrieved from local paths, pasted text, or remote URLs (SKILL.md, Step 1).
- Boundary markers: The instructions do not define protective delimiters or specific ignore directives for the ingested content to prevent instruction override.
- Capability inventory: The skill identifies and modifies project theme CSS files (e.g., globals.css) and reads components.json configuration (SKILL.md, Step 2 and Step 4).
- Sanitization: No validation or sanitization of the markdown prose or token values is performed prior to processing and application.
Audit Metadata