Embed your agent in Claude.ai
Paste a single URL into a Claude conversation. The response is one self-contained HTML document: three.js, GLTFLoader, and your agent's GLB are all inlined. No script blocked by Claude's sandbox CSP, no fetch that ever leaves the page.
Configure
Public agents you can embed
Browse allClaude artifact sandbox CSP
Live preview
Paste-into-Claude snippet
Generate to see the snippet…
Self-contained
three.js, GLTFLoader, and the GLB are all inlined. The page renders with zero network calls, fully compatible with Claude's sandbox CSP that forbids fetch and external scripts.
Sandbox-locked CSP
The response sets a CSP mirroring Claude's own, so behavior in the live preview matches behavior inside Claude.ai. No surprises between staging and prod.
Size budget
Models over 6 MB are rejected, because Claude's artifact panel can hang on huge inlined payloads. Inlining base64-encodes the model, which costs four bytes per three, on top of the flat 565 KB viewer: a 6 MB model lands at roughly 8.6 MB of HTML. That total, not the model, is what the size stat grades.