Convert Stitch designs into production React + Vite dashboards with TanStack Query, accessible tokens from DESIGN.md, and Web3-ready patterns (ethers/viem).
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The REQUIRED runtime workflow ingests outsider-authored free text when it performs `prefix:get_screen` to fetch Stitch screen contents (and then persists/reads the returned HTML), which the LLM uses for component generation; this is public/third-party (Stitch) content not authored by the operating user.
535b088
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.