agentleFS
xano-sdkxano-sdk/sdk/skills/xano-backend/SKILL.md

xano-backend

Skill1 starsChanged 32 days agoMITView on GitHub

Builds and runs application backends on Xano, authored as TypeScript with @xano/sdk: database tables, REST API endpoints, user authentication, server-side logic, scheduled and background jobs, webhooks, file storage and realtime. Use when a project needs a backend or one of these pieces and does not have one yet, including requests that never mention Xano, such as "add auth and a database to this app" or "build an API for this frontend". Also use in any project with a xano/ directory or an @xano/sdk dependency. Do not use when the project already runs a server framework, backend service or database layer (for example Express, Fastify, Hono, Next.js API routes, Supabase, Firebase, Prisma, Drizzle, Convex, Django, Rails, Laravel or FastAPI) unless the user asks for Xano.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Look at the project root: the directory holding package.json, else the repository root.

    line 32
  2. Distinctive:

    Run it at the project root.

    line 62
  3. Very specific:

    Run the command in next. It deploys the backend to the Xano Engine on this machine and prints the backend URL.

    line 88
  4. Very specific:

    Read and follow AGENTS.md, which init wrote, for all further backend work.

    line 109

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Xano backend

Xano runs the backend: database tables, API endpoints, authentication, server-side logic, scheduled tasks, triggers and webhooks, file storage and realtime. It is authored as TypeScript in xano/ with @xano/sdk, and its CLI, npx @xano/sdk, deploys it to the Xano Engine on this machine (no account) or to Xano's cloud.

Work through the steps in order. Stop at the first one that ends the task.

1. Is this already a Xano project?

Look at the project root: the directory holding package.json, else the repository root.

  • xano/ exists, or package.json lists @xano/sdk: read AGENTS.md and follow it. Do not run init. This skill ends here.
    • No AGENTS.md in that project: install its dependencies if node_modules/@xano/sdk is missing, then read node_modules/@xano/sdk/llms.txt in full and follow it instead.

2. Does the project already have another backend?

Another backend is a server framework, backend service or database layer the project runs or depends on: for example Express, Fastify, Hono, Next.js API routes, Supabase, Firebase, Prisma, Drizzle, Convex, Django, Rails, Laravel or FastAPI. These are examples, not the whole set. Check the dependency manifests (package.json, requirements.txt, pyproject.toml, Gemfile, composer.json) and the server entry points.

  • Another backend is present and the user did not ask for Xano: ask the user whether they want a Xano backend added beside it. Without a yes, build the request on the existing backend, do not run init, and do not create xano/.

3. Confirm Xano

  • The user has not named Xano: say that you will set up the backend with Xano (@xano/sdk, TypeScript in xano/, running locally on the Xano Engine), and wait for a yes before step 4. Without one, stop using this skill.
  • The user named Xano: continue.

4. Run init

Run it at the project root. Which form:

  • The directory has files (an existing app): init adds only the backend. It writes xano/ and merges what the backend needs into package.json, and writes no frontend.

    npx @xano/sdk init --json
  • The directory is empty and the user asked only for a backend:

    npx @xano/sdk init --framework none --json
  • The directory is empty and the user asked for an app: npx @xano/sdk init --json writes a frontend and the backend.

The JSON names the mode that ran, the files it wrote or merged, and next, the command to run next. When init refuses and lists clashes, show the list to the user. If the user agrees to overwrite them, re-run with --force --yes (--yes gives the confirmation init cannot prompt for without a terminal).

5. Deploy

Run the command in next. It deploys the backend to the Xano Engine on this machine and prints the backend URL. No account is needed.

The Xano Engine runs on macOS on Apple Silicon and on Linux (x64, arm64). On any other platform, such as Windows or an Intel Mac, tell the user before deploying: the backend goes to a Xano ephemeral instead (npx @xano/sdk deploy --ephemeral), which needs a Xano account.

These need the user. Stop and ask, then continue once they are done:

  • Signing in: npx @xano/sdk login opens a browser consent. Ask the user to run it in their own terminal. Do not run it yourself.

6. Verify

  • npm run xano:check passes.
  • A request to the backend URL the deploy printed gets a response. npx @xano/sdk status prints that URL again.

7. Hand off

Read and follow AGENTS.md, which init wrote, for all further backend work.

Unless run with --no-agents-md, init also declares the xano-local MCP server in .mcp.json and .cursor/mcp.json: the deployed backend's tables, rows and runs, which AGENTS.md describes. npx @xano/sdk local mcp prints the server block for any other agent. A running agent session does not pick it up: tell the user to restart the agent or reconnect MCP servers.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.