Two-command setup
Install the xano-backend skill, then run init. It connects Cursor to the Xano Engine’s local MCP server, so your tables, rows, functions and tests are in reach from the first prompt.
Cursor’s agent writes your backend in TypeScript with the Xano SDK, right in your editor, typed and checked as you go. Run it on the Xano Engine on your machine, then deploy it to Xano: database, APIs and hosting included.
npx @xano/sdk agent-skill install Install the xano-backend skill, then run init. It connects Cursor to the Xano Engine’s local MCP server, so your tables, rows, functions and tests are in reach from the first prompt.
Your backend is typed TypeScript in your project, so autocomplete and type checks cover every line the agent writes. Rename a column and every consumer lights up red.
Run it on the Xano Engine, test it, see what it compiled to in the builder, and promote to production only through a release.
Add the xano-backend skill to each coding agent on your machine, Cursor included. When a project needs a backend and has none, the skill has Cursor’s agent confirm Xano with you, run init, deploy, and follow the AGENTS.md that init wrote.
npx @xano/sdk agent-skill install init writes a TypeScript backend under xano/, a React or Svelte frontend, the xano:deploy scripts and an AGENTS.md brief. It also connects Cursor to the Xano Engine’s local MCP server, where Cursor reads tables and rows, seeds data, and runs functions and tests. Already have an app? Run init inside it and it adds only the backend.
npx @xano/sdk init my-app && cd my-app Ask Cursor’s agent for tables, endpoints, or workflows. It writes TypeScript in your project files, grounded in llms.txt and AGENTS.md, and your editor checks every line. Tab-complete s., f. and input. to discover the whole catalog, and a backend rename is a compile error rather than a 404.
npm run xano:deploy runs the backend on the Xano Engine on your machine and points your frontend at it. Open the builder link it prints to see what your code compiled to. Then npm run xano:deploy:ephemeral for a live URL on Xano’s cloud, and a release to reach your workspace.
npm run xano:deploy Cursor’s agent works inside your editor, so the backend is type-checked while it is written, not after. A short project rule keeps the agent on the AGENTS.md brief that init wrote.
Keep the brief in one place. A project rule in .cursor/rules/ scoped to xano/ tells Cursor’s agent to follow AGENTS.md and the SDK’s llms.txt, so the same brief serves Cursor and every other agent on the project.
Every kind, statement, filter and field type is typed. Tab-complete s., f., c., fl. and input. to find the whole catalog, and Cursor’s agent sees the same types you do.
A generated route manifest types your frontend’s calls, so renaming a backend column turns every consumer red: a compile error rather than a 404. Fix it in Cursor, then npm run xano:deploy.
init connects Cursor to the Xano Engine’s local MCP server, so the agent reads tables and rows, seeds data, and runs functions and tests without leaving the editor.
.cursor/rules/xano.mdc---
description: Xano backend written with the Xano SDK
globs: xano/**
alwaysApply: false
---
Follow AGENTS.md (npx @xano/sdk init wrote it)
for every change under xano/.
- Write the backend in TypeScript with @xano/sdk.
- Look up the API in the SDK's llms.txt. Use
autocomplete on s. f. c. fl. input. before guessing.
- Fix every type error before you deploy.
- After a change: npm run xano:deploy,
then npx xanosdk test run-all. AI can generate apps fast. Xano can help you make them production-ready, reliable, and built to last.
How Xano governs AI-generated codeEvery command on this page comes from the Xano SDK. See how coding agents build on Xano →
See what your code compiled to as a Function Stack in Canvas View: transparent, auditable, human-readable.
Production-ready PostgreSQL with triggers, relationships, and direct access. No migrations, no DevOps.
Instant REST APIs with auth, rate limiting, and custom logic. Ready to connect to any frontend or service.
Describe your backend. AI builds schemas, APIs, auth, and logic, governed from the start.
Write the backend as typed TypeScript in your repo, from any IDE or agent, with full governance intact.
Request tracing, performance dashboards, and audit logs across every system.
Branch, test, and promote. Git-based versioning so nothing untested reaches production.
HIPAA, SOC 2, GDPR, and ISO compliance with fine-grained access control. Built in from day one.
Run npx @xano/sdk agent-skill install to add the xano-backend skill to Cursor (or npx skills add xano-sdk/sdk), then npx @xano/sdk init my-app. init connects Cursor to the Xano Engine’s MCP server; npx xanosdk local mcp shows how an agent connects.
Yes. Run init inside the existing project and it adds only the backend: xano/, the xano:deploy scripts and AGENTS.md. If the backend already lives in a Xano workspace, init --from workspace reads it back as TypeScript, so Cursor’s agent edits your real endpoints instead of new ones.
Yes. Every kind, statement, filter and field type is typed, so your editor’s autocomplete is the fastest lookup: tab-complete s., f., c., fl. and input.. A generated route manifest types your frontend’s calls too, so a backend rename is a compile error rather than a 404.
The TypeScript compiler flags it in your editor, before you deploy. The tests you author run with npx xanosdk test run-all, and changes go to the Xano Engine or a disposable ephemeral first, so you test them and see what they compiled to before anything goes live.
No, but a short one helps. init writes AGENTS.md, the brief every agent follows. A project rule in .cursor/rules/ scoped to xano/ can point Cursor’s agent at it, so Claude Code, Codex and Cursor all build to the same brief.
Edit it like any other file. The TypeScript lives in xano/ in your project, so you change it in Cursor, let the compiler check it, and redeploy. Canvas View in the Xano editor shows your team what it compiled to. Everything stays visible, editable and portable.
Write it in Cursor. Run it on Xano.