Open-source platform for 3D AI agents. Turn text or a photo into a rigged, animated GLB avatar, give it an LLM brain, memory and a wallet, and embed it anywhere with one web component. Remote MCP server, x402 payments, Three.js, browser-native.
JavaScript 3d-generationai-agentsanimationavatardigital-humangenerative-aiglbgltfimage-to-3dllmmcpmcp-servermodel-context-protocolsolanatext-to-3dthreejsweb-componentswebglwebxrx402
# Docker build context excludes. The worker Dockerfiles build from the repo root
# and install their own deps inside the image, so the host's node_modules (multi-GB)
# must never be uploaded to the daemon. Secrets and local-only artifacts are
# excluded too — they must never enter a build context.
**/node_modules
.git
.env
.env.*
!.env.example
*.log
npm-debug.log*
.DS_Store
.vercel
coverage
playwright-report
test-results
*.local
# Secrets and local dumps — never bake into an image. The root holds live config
# JSON (vercel.json, cors.json, server*.json) the runtime needs, so exclude by
# specific name/glob, never a blanket *.json.
.x402-ring-secrets.json
*-log-export-*.json
three.ws-log-export-*.json
*.pem
*.key
# Worker signing-key caches. A worker Dockerfile COPYs its own directory wholesale,
# so a local .cache/ of raw secret keys (workers/agora-citizens/.cache holds one
# devnet keypair per citizen) would be baked into the pushed image AND would make
# the deployed fleet drive the same on-chain identities as the local one. They are
# gitignored; they must be build-ignored too. The worker regenerates them at boot.
workers/*/.cache