Tanstack Start logo

Tanstack Start

OrganizationPopular
vercel-labs
tanstack-start

Build JSON-defined TanStack Start applications with @json-render/tanstack-start. Use for Start route specs, splat routing, SSR loaders, head metadata, layouts, and client navigation. Do not use for generic TanStack Router apps that do not use json-render.

Overview

Publishervercel-labs
Repositoryjson-render
Skill nametanstack-start
Stars
16.5K
Forks
887
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by vercel-labs on GitHub. Read the source before you install it.

Installation

Install the Tanstack Start AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/vercel-labs/json-render.git /tmp/json-render
mkdir -p .claude/skills
cp -r /tmp/json-render/skills/tanstack-start .claude/skills/tanstack-start
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Tanstack Start in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Tanstack Start on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Tanstack Start is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

@json-render/tanstack-start

Use this integration when a TanStack Start app needs complete pages or routes described by json-render specs.

Install

bash
npm install @json-render/core @json-render/react @json-render/tanstack-start

Application Spec

Include the server-safe built-in definitions in the generation catalog. The renderer supplies their component implementations.

typescript
import { defineCatalog } from "@json-render/core";
import {
  schema,
  startComponentDefinitions,
} from "@json-render/tanstack-start/server";

const catalog = defineCatalog(schema, {
  components: {
    ...startComponentDefinitions,
    Card: cardDefinition,
    Shell: shellDefinition,
    Navigation: navigationDefinition,
    Home: homeDefinition,
    Post: postDefinition,
  },
  actions: {},
});

Then use StartAppSpec and TanStack Router route patterns:

typescript
import type { StartAppSpec } from "@json-render/tanstack-start";

export const spec: StartAppSpec = {
  metadata: {
    title: { default: "Site", template: "%s | Site" },
  },
  layouts: {
    main: {
      root: "shell",
      elements: {
        shell: { type: "Shell", props: {}, children: ["nav", "slot"] },
        nav: { type: "Navigation", props: {}, children: [] },
        slot: { type: "Slot", props: {}, children: [] },
      },
    },
  },
  routes: {
    "/": {
      layout: "main",
      metadata: { title: "Home" },
      page: {
        root: "home",
        elements: {
          home: { type: "Home", props: {}, children: [] },
        },
      },
    },
    "/posts/$slug": {
      layout: "main",
      loader: "post",
      staticParams: [{ slug: "hello" }],
      page: {
        root: "post",
        elements: {
          post: {
            type: "Post",
            props: { value: { $state: "/post" } },
            children: [],
          },
        },
      },
    },
  },
};

Routes use /posts/$slug for named parameters and /docs/$ for a splat. Splat loader parameters are slash-delimited strings under _splat. Escape route-key slashes as ~1 when generating RFC 6902 patches.

Every layout needs a Slot element. Declare Slot and Link through startComponentDefinitions; do not require consumers to register React implementations for them.

Server Helpers

typescript
import { createStartApp } from "@json-render/tanstack-start/server";

export const { getPageData, getHead, getStaticPaths } = createStartApp({
  spec,
  loaders: {
    post: async ({ slug }) => ({ post: await getPost(slug as string) }),
  },
});

State merge precedence is application state, layout state, page state, then loader data. getHead merges app and route metadata into TanStack meta and links descriptors. getStaticPaths includes static routes plus dynamic routes with staticParams. Convert its strings to { path } objects for TanStack Start's top-level pages plugin option. Loader params are URL-decoded, while values from staticParams are URL-encoded in generated paths. Route matching treats trailing slashes as optional and accepts encoded or decoded pathname representations so loader data and metadata resolve the same static route.

Route Wiring

tsx
import { createFileRoute, notFound } from "@tanstack/react-router";
import {
  PageRenderer,
  StartErrorBoundary,
  StartLoading,
  StartNotFound,
} from "@json-render/tanstack-start";
import { getHead, getPageData } from "@/lib/json-app";

export const Route = createFileRoute("/$")({
  loader: async ({ location }) => {
    const data = await getPageData({ pathname: location.pathname });
    if (!data) throw notFound();
    return data;
  },
  head: ({ match }) => getHead({ pathname: match.pathname }),
  component: () => <PageRenderer {...Route.useLoaderData()} />,
  pendingComponent: StartLoading,
  errorComponent: StartErrorBoundary,
  notFoundComponent: StartNotFound,
});

TanStack Router loaders run on both the server and client. If a spec factory or named loader uses credentials, database clients, or server-only imports, wrap getPageData and getHead in a TanStack Start createServerFn; do not import that server code directly into an isomorphic route loader.

Root Provider

Wrap the root route's Outlet with StartAppProvider. Render HeadContent so metadata from getHead reaches the document.

tsx
import {
  createRootRoute,
  HeadContent,
  Outlet,
  Scripts,
} from "@tanstack/react-router";
import { StartAppProvider } from "@json-render/tanstack-start";
import { spec } from "@/lib/spec";

export const Route = createRootRoute({
  component: () => (
    <html lang="en">
      <head>
        <HeadContent />
      </head>
      <body>
        <StartAppProvider
          registry={registry}
          handlers={handlers}
          spec={spec}
        >
          <Outlet />
        </StartAppProvider>
        <Scripts />
      </body>
    </html>
  ),
});

Use StartLoading, StartErrorBoundary, and StartNotFound for TanStack Router's pendingComponent, errorComponent, and notFoundComponent options. When StartAppProvider receives spec, each component selects the matched route's corresponding fallback. Explicit fallback props override that lookup. Pass named $computed implementations through StartAppProvider.functions. The default error boundary invalidates the router and reruns a failed loader when the user selects Try again.

Import React components from @json-render/tanstack-start. Import schema, createStartApp, matchRoute, resolveMetadata, and static path helpers from @json-render/tanstack-start/server.

Frequently asked questions

What does the Tanstack Start AI skill do?

Build JSON-defined TanStack Start applications with @json-render/tanstack-start. Use for Start route specs, splat routing, SSR loaders, head metadata, layouts, and client navigation. Do not use for generic TanStack Router apps that do not use json-render.

Why use Tanstack Start on TypingMind?

Because you install it once and use it with any model. Tanstack Start is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Tanstack Start in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/vercel-labs/json-render/tree/main/skills/tanstack-start. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Tanstack Start?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Tanstack Start?

As many as you like. As long as a model supports skills, you can use Tanstack Start with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Tanstack Start AI skill free?

Yes. It is published on GitHub by vercel-labs under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇