stitch-react-components
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interprets external design data (HTML and screenshots), which creates a surface for indirect prompt injection where malicious instructions could be embedded in the design source.
- Ingestion points: External data is ingested via
scripts/fetch-stitch.shand processed from the.stitch/directory. - Boundary markers: The instructions do not define explicit delimiters or security boundaries to isolate design data from the agent's core instructions.
- Capability inventory: The skill possesses file-writing capabilities (creating TSX and TS files) and network access for asset retrieval.
- Sanitization: The skill lacks security-focused sanitization of the ingested HTML content before using it for code generation.
- [EXTERNAL_DOWNLOADS]: The skill uses a shell script
scripts/fetch-stitch.shto download assets viacurl. While this is an essential part of the design-to-code workflow, it involves network requests to dynamically provided URLs.
Audit Metadata