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
// An in-process Postgres (PGlite) behind the same tagged-template `sql` shape
// api/_lib/db.js exports, for tests that must exercise real SQL semantics:
// check constraints, partial unique indexes, ON CONFLICT inference, jsonb
// operators. Mock api/_lib/db.js with `{ ...actual, sql: db.sql }` and create
// only the tables the code under test touches.
//
// Like the production wrapper, a query is lazy (it runs when awaited) and an
// interpolated sql`...` fragment is inlined with its placeholders renumbered,
// so modules that build column lists or WHERE clauses as fragments work here.
import { PGlite } from '@electric-sql/pglite';
const FRAGMENT = Symbol('pglite-sql-fragment');
function compile(fragment, params) {
let text = '';
fragment.strings.forEach((part, i) => {
text += part;
if (i >= fragment.values.length) return;
const v = fragment.values[i];
if (v && v[FRAGMENT]) {
text += compile(v, params);
} else {
params.push(v instanceof Date ? v.toISOString() : v);
text += `$${params.length}`;
}
});
return text;
}
export function createPgliteSql() {
const state = { pg: new PGlite(), calls: [] };
const run = async (fragment) => {
const params = [];
const text = compile(fragment, params);
state.calls.push(text);
const out = await state.pg.query(text, params);
return out.rows;
};
const sql = (strings, ...values) => {
const fragment = { [FRAGMENT]: true, strings, values };
let promise = null;
const settle = () => (promise ??= run(fragment));
fragment.then = (onOk, onErr) => settle().then(onOk, onErr);
fragment.catch = (onErr) => settle().catch(onErr);
fragment.finally = (fn) => settle().finally(fn);
return fragment;
};
return {
sql,
exec: (ddl) => state.pg.exec(ddl),
query: (text, params) => state.pg.query(text, params).then((r) => r.rows),
calls: state.calls,
};
}