WildflowerJS Reactive JS, No BS*

A no-build reactive JavaScript framework, rooted in the web platform.
No build step. No dependencies. No compromises.

Latest release: v1.5.0 · see what's new
<script src="wildflower.min.js"></script> ...and start building.

Back to Basics

With WildflowerJS, you write 100% standard code. HTML stays HTML. JavaScript stays JavaScript. CSS stays CSS. There's no JSX, templating language, or custom syntax to learn. If you know the standards, you already know how to use WildflowerJS.

WildflowerJS extends the web platform. It doesn't replace it.

Your Development Simplified

Because you develop with 100% web standards, every tool in your existing chain already understands the code: IDE, browser DevTools, linter, formatter, screen reader, SEO crawler. There's nothing to install, and no custom file types or sourcemaps. Save the file, refresh, and your change is live.

Just be a web developer.

Batteries Included: One Mental Model

Router, SSR, queries, stores, computed properties, two-way binding, event modifiers, data pools, and TypeScript types, all built in, all using the same API. Learn data-bind once and you know binding everywhere: in lists, pools, stores, plugins. There's no five-library stack to keep in sync.

One script tag. Everything you need.

<div data-component="counter">
  <span data-bind="count"></span>
  <button data-action="increment">
    +1
  </button>
</div>

<script>
wildflower.component('counter', {
  state: { count: 0 },
  increment() { this.count++ }
})
</script>

How It Works

data-bind connects state to the DOM.

data-action connects events to methods.

this.count++ triggers a precise DOM update.

Mutate state. The DOM updates.

Two Reactivity Modes

data-list is for automatic reactivity: mutate state, the DOM updates. data-pool is for explicit control: plain objects, zero proxy overhead, you say what changed.

Both use the same template syntax and differ in performance profile, from interactive forms to per-frame particle systems. You choose the tradeoff that fits the job.

Try it. Right-click, inspect this demo. Every dot is a real DOM element.

See full demo →

* Build Step

No Toolchain

Modern frameworks ask you to install a compiler, a bundler, a package manager, hundreds of fragile transitive dependencies, and a framework-specific file format, before you write a single line of your application.

WildflowerJS was built starting from a single principle: no build step, no tooling. Ever.

WildflowerJS asks you to add a script tag.

There's no CLI scaffolding step, config file, or .vue/.jsx/.svelte source format. You don't debug through sourcemaps or wait on a build pipeline. Your project has zero dependencies.

Performance isn't a tradeoff. Build steps optimize bundle delivery, not the runtime work that follows it. WildflowerJS writes directly to the DOM, with no virtual DOM or reconciliation pass between state change and update, so it doesn't need a build step to be fast.

The framework is full-featured without the toolchain, including router, SSR, stores, computed properties, transitions, and pools.

my-app/
  index.html
  app.js
  style.css
  wildflower.min.js

That's the entire project. No package.json, no node_modules, no config files. NONE of that.

No Install. No Attack Surface.

Every dependency you install lets a maintainer you have never met run scripts on your dev machine and in your CI. A typical React + Vite + UI‑lib setup pulls in 300+ transitive packages before you write a feature.

Each one is a potential intrusion vector. NPM worms, OAuth chains compromising deploy platforms, postinstall hijacking: the supply chain is now where production code gets compromised, not the deploy. And signing isn't a backstop: Mini Shai‑Hulud (May 2026) compromised 170+ packages whose malicious versions carried valid SLSA Build Level 3 provenance, because the attestation came from build infrastructure the worm had already taken over.

A WildflowerJS project has none of that surface. There is no npm install, postinstall script, or transitive package graph. The framework is one file you copy or pin by hash.

As of v1.1, the same holds for building the framework itself. WildflowerJS bundles with a vendored rollup and terser pipeline pulled as three SHA‑512‑pinned tarballs, with no npm install and no transitive packages in the build path. The entire toolchain is three files verified by hash.

A typical React/Vue project:

  npm install
  ├── hundreds of packages
  ├── from hundreds of maintainers
  ├── postinstall scripts run on install
  └── tens to hundreds of MB of transitive code

WildflowerJS:

  <script src="wildflower.min.js"></script>
  └── 1 file.
      No transitive dependencies.

No Compromise

WildflowerJS doesn't compromise performance for ease-of-use. Even with no build step, WildflowerJS performs at the level of frontier frameworks on the official js-framework-benchmark board, where its data-pool entry outpaces every major framework and its standard entry sits with the fastest signal-based compilers. And for per-frame workloads, data pools lead every framework we tested in our Lorenz attractor simulation demo.

The charts here are the overall geomean standings and the operation breakdown from the official September 2026 run, plus the sustained frame rate from our per-frame animation sweep. Click any chart to see it full size.

Delivery is fast too, because there's less to deliver. It ships as one file, with no runtime split across chunks and no hydration pass. Lighthouse scores hold their own against compiled frameworks without a single build artifact.

WildflowerJS doesn't trade simplicity of interface for performance of implementation.

Benchmark setup: the two js-framework-benchmark charts show the official September 2026 run (Chrome 152; MacBook Pro 14, M4 14/20 cores, 48 GB RAM, macOS 26.6.2; puppeteer driver), operations 1 through 9, total-duration medians, lower is better. The frame-rate chart is our own sweep: each framework's fastest variant on the Lorenz attractor for 8 seconds per particle count, fullscreen on a 120 Hz panel, higher is better; Apple M5 Pro, 24 GB RAM, macOS 26.5.2, Google Chrome 150 (stable, headed).

Bar chart of the official weighted geometric mean slowdown versus the fastest implementation per operation, Chrome 152: WF-pool 1.09, Vue Vapor 1.12, Solid 1.13, WF 1.16, Svelte 1.17, Vue 1.31; vanilla 1.04 and React 1.58 not shown. Lower is better.
Geomean slowdown vs fastest per operation. Lower is better.
Grouped bar chart of all nine js-framework-benchmark operations for Solid, Svelte, Vue, Vue Vapor, WF, and WF-pool from the official Chrome 152 run, with per-operation rankings. WF-pool is fastest on most operations.
All nine operations, side by side. Stars mark the fastest.
Line chart of sustained FPS versus particle count on the Lorenz attractor for Solid, Svelte, Vue, Vue Vapor, WF, and WF-pool. WF-pool holds the highest frame rate at every count, staying above 60 FPS past 4500 particles.
Per-frame animation. Sustained FPS as particle count grows; higher is better.

No Lock-in

WildflowerJS works with the DOM, not instead of it. There's no virtual DOM intercepting your code and no compiler rewriting your markup. The render cycle is yours alone.

That means Leaflet, DataTables, Chart.js, D3, Three.js, any library that touches the DOM, just works. There are no wrapper packages or framework-specific escape hatches required. Drop in a script tag, it's ready to go.

Because your code is standard HTML and JavaScript, you're never locked in. Your skills transfer and your code is more portable. If you outgrow the framework, your knowledge doesn't expire.

This also means your "ecosystem" is the whole of vanilla JS, with no compromises or hacks.

<!-- Use any library directly -->
<div data-component="map-view">
  <div id="map" style="height: 400px"></div>
</div>
wildflower.component('map-view', {
  state: { lat: 51.505, lng: -0.09 },
  init() {
    // Leaflet works as-is. No wrappers.
    this._map = L.map('map')
      .setView([this.lat, this.lng], 13);
    L.tileLayer('https://{s}.tile.osm.org'
      + '/{z}/{x}/{y}.png').addTo(this._map);
  }
})

Precise Reactivity

When you write this.count++, WildflowerJS updates the single DOM node bound to count. Nothing else is touched. There's no tree diffing or reconciliation pass to figure that out.

You get fine-grained updates and a simple mental model. Change a property, the bound element updates. That's the entire reactivity model.

Other frameworks ask you to learn signals, accessors, memos, effects, and subscription lifecycles to achieve what WildflowerJS does with a standard JS property assignment.

wildflower.component('dashboard', {
  state: {
    users: 1420,
    status: 'healthy'
  },
  computed: {
    summary() {
      return this.users + ' users, ' + this.status;
    }
  },
  refresh() {
    this.users = 1421;
    // Only the elements bound to 'users'
    // and 'summary' update. Everything
    // else on the page is untouched.
  }
})

One Reactivity Model. Everywhere.

Components, Stores, and Plugins, Pools, and now Data Queries all share the same reactive foundation. State, computed properties, and methods work identically no matter where they live. Learn it once, it works the same way across all of those entities.

Other frameworks make you learn a different system for each layer. React components use hooks, but stores need Redux or Zustand, which are completely different APIs. Vue components use reactive data, but Pinia stores have their own patterns. Every layer is a new mental model.

In WildflowerJS, there's one model. A store is a component without a template. A plugin is an entity that extends the framework itself, adding directives, lifecycle hooks, and services. The same this.count++ triggers the same reactivity everywhere.

This makes patterns possible that other frameworks cannot express. A store can run headless physics simulations with tick(), feeding data into a component that renders it through a pool, all using the same reactive primitives, no glue code required.

// Component: reactive UI
wildflower.component('cart', {
  state: { items: [] },
  computed: {
    total() { return this.items.length; }
  }
})

// Store: global shared state
wildflower.store('user', {
  state: { name: '', role: 'guest' },
  computed: {
    isAdmin() { return this.role === 'admin'; }
  }
})

// Plugin: extends the framework
wildflower.plugin({
  name: 'notifications',
  state: { items: [], unreadCount: 0 },
  computed: {
    hasUnread() { return this.unreadCount > 0; }
  },
  add(msg) { this.items.push(msg); this.unreadCount++; }
})
// Access globally: wildflower.$notifications.add(...)

// Same state. Same computed. Same methods.

Live Server Data: Built In, Stays True

With WildflowerJS SSR, the page arrives with its data already in the HTML. The server (your server, whatever back-end you prefer) renders your data into real HTML, so the first paint is real content, indexable and readable before a line of JavaScript runs. And because the markup is genuine HTML, hydration reads the page's state straight back out of the document. Server-rendered components end up exactly equivalent to client-rendered ones.

v1.3 brings data-query, which does for the rest of the page's life what SSR does for first load. Most frameworks hand you fetch() and leave the rest to you. There's an entire ecosystem of client data libraries that exists to fill that gap. WildflowerJS makes it a declaration instead. Name a source, point an element at it, say how fresh it should stay. Loading and error states, refresh on demand, request racing, and the whole refresh ladder (poll, conditional GET, focus, reconnect, server push) come with it. There is also no query language. Refinement is an ordinary computed property, and filtering happens client-side without a network round trip.

v1.5 completes the shape with writes. A query that declares where its rows come from can declare where changes go: to: is the transport, write() applies the change on screen immediately, and confirmation: decides what the server's answer means. If the server refuses, only the fields that write still owns revert, so two writes to the same row never clobber each other and you write no cancellation logic to get it. Computed properties may also return a promise now, holding the last settled value while the next one resolves.

Together, Wildflower's SSR and data-query cover one job at two different times. The server renders the page with real data. Because hydration reads the page itself, there's no flash of empty content, no loading spinner over data the user can already see, and no hydration scripts locking up the main thread. The server's render is the actual UI. When paired with data-query, your SSR becomes the first result of a standing query. The query adopts that markup and keeps it updated from there.

In the example above, the markup is 100% HTML.

<div data-component="product-board">
  <p data-show="$products.isLoading">
    Loading…
  </p>
  <p data-show="$products.error">
    Failed.
    <button data-action="retry">Retry</button>
  </p>

  <span data-bind="$products.count"></span>
  products

  <tbody data-query="products">
    <template>
      <tr>
        <td data-bind="name"></td>
        <td data-bind="stock"></td>
      </tr>
    </template>
  </tbody>
</div>
// The entire data layer:
wildflower.query('products', {
  from: '/api/products',
  key: 'id',
  refresh: ['focus', 'etag:60'],

  // v1.5: where changes go
  to: '/api/products/:id',
  body: (item) => item,
  confirmation: (d) => d.product
});

// The key plus only what changed. On screen
// at once; if the server refuses, only those
// fields revert.
getQuery('products')
  .write({ id: 42, stock: 40 });

// Server-rendered page? Add data-ssr="true"
// and the markup the server sent becomes the
// query's first result. Live from there.

Data Pools

Every framework wraps collection items in reactive proxies, whether the item needs it or not. WildflowerJS gives you a choice: data-list for push reactivity (automatic), data-pool for pull reactivity (explicit control, zero proxy overhead).

Pools render plain objects with the same template syntax as lists. Mutate the object, call markDirty(), and only that item updates. Full CRUD, selection, bulk operations, all faster than the push-reactive path.

And because pools use pull-based rendering, they scale to simulations, games, particle systems, and data visualizations at native frame rate, which a virtual DOM cannot sustain. No other framework offers this choice.

<div data-component="user-table">
  <tbody data-pool="users" data-key="id">
    <template>
      <tr>
        <td data-bind="name"></td>
        <td data-bind="status"
            data-bind-class="status === 'active'
              ? 'badge success'
              : 'badge inactive'"></td>
      </tr>
    </template>
  </tbody>
</div>
wildflower.component('user-table', {
  pools: { users: {} },

  init() {
    // Populate: plain objects, no proxies
    data.forEach(u => this.pools.users.add(u));
  },

  // Optional: add tick() and the same pool
  // renders every frame. Same template, same
  // data, different rendering frequency.
  // That's the only difference between a
  // display table and a particle system.
})

Built for AI-Assisted Development

Because WildflowerJS is standard HTML and JavaScript, AI code assistants already know how to write it. There's no custom syntax to hallucinate or compiler quirks to work around. The code an AI generates runs exactly as written, with no build step between generation and execution.

WildflowerJS ships an AI-optimized reference page with patterns, anti-patterns, and examples designed for code generation context windows. Our llms.txt file follows the llms.txt convention for machine-readable documentation.

You: "Build me a todo app with
WildflowerJS"

AI reads llms.txt or ai-assistant.html
     ↓
Generates standard HTML + JS
     ↓
<div data-component="todo-app">
  <input data-model="newItem">
  <button data-action="addItem">
    Add
  </button>
  <ul data-list="items">
    <template>
      <li data-bind="text"></li>
    </template>
  </ul>
</div>
     ↓
Open in your browser. It works, and you can read and understand the code.

Async Computed Properties v1.5+

A computed property may return a promise. Bindings keep showing the last good value while the request runs, then update when the new value arrives.

💡 Key Concept: There is nothing new to learn here. Same computed: block, same data-bind, same dependency tracking. As of v1.5 the restriction that a computed must return synchronously is gone.

Return a Promise

Write the computed the way you would write any derived value, and return the request. async functions work too, since an async function returns a promise:

<div data-component="async-user-card">
    <p class="text-muted small mb-2">Each lookup below runs against a simulated 700ms request.</p>

    <div class="mb-3">
        <button class="btn btn-primary btn-sm me-2" data-action="loadUser" data-id="1">Ada</button>
        <button class="btn btn-primary btn-sm me-2" data-action="loadUser" data-id="2">Grace</button>
        <button class="btn btn-primary btn-sm" data-action="loadUser" data-id="3">Edsger</button>
    </div>

    <!-- No value yet means the first load is still running -->
    <p class="text-muted" data-show="!user">Loading directory…</p>

    <div class="card" data-show="user">
        <div class="card-body">
            <h5 class="card-title" data-bind="user.name"></h5>
            <p class="card-text mb-1" data-bind="user.role"></p>
            <p class="card-text text-muted mb-0" data-bind="user.quote"></p>
        </div>
    </div>
</div>
const TEAM = {
    1: { name: 'Ada Lovelace', role: 'Analytical engines', quote: 'The engine weaves algebraic patterns.' },
    2: { name: 'Grace Hopper', role: 'Compilers', quote: 'It is easier to ask forgiveness than permission.' },
    3: { name: 'Edsger Dijkstra', role: 'Algorithms', quote: 'Simplicity is a great virtue.' }
}

// Stands in for fetch('/api/team/' + id).then(r => r.json())
function fetchMember(id) {
    return new Promise(resolve => setTimeout(() => resolve(TEAM[id]), 700))
}

wildflower.component('async-user-card', {
    state: { userId: 1 },

    computed: {
        // Returns a promise. The binding shows the previous member
        // while the new request runs, then updates when it lands.
        user() {
            return fetchMember(this.userId)
        }
    },

    loadUser(event) {
        this.userId = Number(event.target.dataset.id)
    }
})
Live Preview

Watch what happens when you switch members. The card keeps showing the current person for the ~700ms the new request takes, then swaps. No spinner flashes over real content, and nothing in the template had to account for the request at all.

First Load and Loading States

On the very first evaluation there is no previous value, so the computed reads undefined while the request runs. The absence of a value is the loading signal, and the syntax for it already exists:

<h1 data-bind="user.name"></h1>
<p data-show="!user">Loading…</p>

Once a value has arrived, a refresh keeps it on screen while the replacement is in flight, so !user stays false and the loading message never reappears over real data. A loading state that only shows when there is nothing better to show is the behavior you want on both counts, and it costs no new vocabulary.

Inputs Can Change Mid-Flight

When a dependency changes while a request is still running, the computed re-runs and launches a new request. The last call wins. An earlier response that arrives late is discarded, so bindings never go backwards:

<div data-component="async-search">
    <input type="text" class="form-control mb-2" data-model="query"
           placeholder="Search languages (try typing 'ru' quickly)">

    <p class="text-muted" data-show="!results">Loading the index…</p>

    <ul class="list-group" data-list="rows" data-key="name">
        <template><li class="list-group-item py-1" data-bind="name"></li></template>
    </ul>
</div>
const LANGUAGES = ['C', 'C++', 'Clojure', 'Elixir', 'Erlang', 'Go', 'Haskell',
    'Java', 'JavaScript', 'Julia', 'Kotlin', 'Lisp', 'Lua', 'OCaml', 'Pascal',
    'Perl', 'PHP', 'Prolog', 'Python', 'R', 'Ruby', 'Rust', 'Scala', 'SQL',
    'Swift', 'TypeScript', 'Zig']

// Broad queries return more rows, so they take LONGER. Type 'ru' quickly:
// the 'r' request and the 'ru' request overlap, the 'ru' response arrives
// first, and the late 'r' response is discarded. The list always ends up
// matching what you typed last.
function searchLanguages(q) {
    const hits = LANGUAGES
        .filter(l => l.toLowerCase().includes(q.toLowerCase()))
        .map(name => ({ name }))
    const latency = 150 + hits.length * 25
    return new Promise(resolve => setTimeout(() => resolve(hits), latency))
}

wildflower.component('async-search', {
    state: { query: '' },

    computed: {
        // Async: re-runs on every keystroke, superseding the request in flight
        results() {
            return searchLanguages(this.query)
        },

        // Ordinary sync computed chained off the async one. Downstream
        // computeds need no special handling.
        rows() {
            return this.results || []
        }
    }
})
Live Preview

The previous results stay on screen while you type, and the out-of-order responses sort themselves out. The component code contains no request bookkeeping, debounce, or cancellation logic.

What a Binding Sees, Moment by Moment

Moment What a binding sees
First evaluation, request in flight undefined. Render a first-load state with data-show="!user".
Refresh in flight The previous value, unchanged.
The request lands The new value. Bindings update once.
A newer run started meanwhile Nothing. The superseded response is discarded silently.
The promise rejects undefined, and the entity's onError hook receives the error.

Read Reactive Inputs Before the Request

Dependency tracking is synchronous. Reads that happen during the computed's own run are tracked; reads inside a .then() callback or after an await happen later, on nobody's watch. Read every reactive input up front, then build the request from locals:

computed: {
    // ✅ this.userId is read synchronously, so changing it re-runs the computed
    user() {
        const id = this.userId
        return fetch('/api/users/' + id).then(r => r.json())
    },

    // ✅ Same rule with async syntax: reads BEFORE the first await are tracked
    async profile() {
        const id = this.userId
        const res = await fetch('/api/profiles/' + id)
        return res.json()
    },

    // ❌ this.userId is read after the await, so it is never tracked and
    // the computed will not re-run when it changes
    async stale() {
        await somethingElse()
        return fetch('/api/users/' + this.userId).then(r => r.json())
    }
}

This is the same rule that governs conditional reads in sync computeds. The tracker records what actually executes during evaluation, and an await ends that window.

Errors

A rejected promise puts the computed in the same errored state a thrown sync computed reaches. Bindings read undefined, and the entity's onError hook receives the rejection:

wildflower.component('user-panel', {
    state: { userId: 1 },

    computed: {
        user() {
            return fetch('/api/users/' + this.userId).then(r => {
                if (!r.ok) throw new Error('HTTP ' + r.status)
                return r.json()
            })
        }
    },

    onError(error, context) {
        // Every failure in this component arrives here, not just this
        // computed's, so claim only the one this branch understands.
        if (context.lifecycle === 'computed' && context.computedName === 'user') {
            console.warn('user load failed', error)
            return true    // handled; stops propagation to parent boundaries
        }
        return false       // everything else continues to a parent boundary
    }
})

onError is the entity's one error hook, so init failures, action handlers, and every other computed report to the same function. Read context.lifecycle and context.computedName to identify what failed before handling it. Returning true marks the error handled and stops propagation; returning false passes it to a parent error boundary. Returning nothing at all counts as handled, so a handler that falls off the end swallows every error the component can raise. See Error Boundaries for the full contract.

Because the errored computed reads undefined, the same data-show="!user" element that covered the first load doubles as the empty state after a failure. The next successful run replaces it, so recovery needs no extra code either. A change to any dependency relaunches the request.

That hook only sees the newest request. When a dependency change relaunches while an earlier request is still in flight, the superseded request is discarded whole, so a late rejection from it never reaches onError, the same way a late resolution never reaches the binding. A failure of a request nobody is waiting on is not an application error.

Stores and Plugins Work the Same Way

Components, stores, and plugins share one computed implementation, so a store computed may return a promise too. This is often the better home for a request several components read:

wildflower.store('directory', {
    state: {},
    computed: {
        members() {
            return fetch('/api/team').then(r => r.json())
        }
    }
})

wildflower.component('team-list', {
    subscribe: ['directory'],
    computed: {
        rows() {
            return this.stores.directory.members || []
        }
    }
})

The store owns the request; every subscribed component reads the resolved value through ordinary computed chaining. One fetch serves the page.

Item-Level Computeds Stay Synchronous

A computed that takes the item as a parameter, like label(item), runs once per list row through the list renderer rather than through the entity's reactive graph. That per-row path does no async tracking, so a promise returned there would bind as text, and a list of a thousand rows would issue a thousand uncoordinated requests. Dev builds warn with WF-235 when an item-level computed returns a promise.

computed: {
    // ❌ One request per row; dev builds warn (WF-235)
    avatar(item) {
        return fetch('/api/avatars/' + item.id).then(r => r.json())
    },

    // ✅ One request for the collection
    avatars() {
        return fetch('/api/avatars?ids=' + this.state.items.map(i => i.id).join(','))
            .then(r => r.json())
    },

    // ✅ Derive the row array from the landed collection, and bind the
    // list to it: data-list="rows"
    rows() {
        const all = this.avatars || {}
        return this.state.items.map(i => ({ ...i, avatar: all[i.id] || null }))
    }
}

When avatars resolves, rows recomputes through ordinary chaining and the keyed list reconciler updates the rows in place. The live search demo above uses the same shape.

How It Works

When a computed returns a promise, the framework holds the node at its previous value, remembers which run launched the request, and attaches a continuation. When the promise settles, the settled value wakes the computed's observers through the same dependency tracking every other change uses. The computed's body does not re-run on resolution, so a fetch never re-triggers itself, and a response from a superseded run fails its generation check and is dropped. The reactive graph itself never learns async exists, which is why everything downstream, from chained computeds to data-list, keeps working unmodified.

Best Practices

✅ Do
  • Return the promise itself (or use an async function)
  • Read every reactive input before the first await or .then()
  • Use data-show="!value" for the first-load state
  • Put shared requests in a store computed
  • Handle rejections in onError
❌ Don't
  • Write state from inside the computed; return the value instead
  • Return promises from item-level computeds (WF-235)
  • Add manual cancellation or debounce for correctness; supersession already guarantees the last call wins
  • Invent a loading flag in state; the absence of the value is the signal
💡 When to reach for data queries instead: An async computed derives one value from reactive inputs. When you need freshness policies, retry, optimistic writes, or list reconciliation on top of the fetch, that lifecycle machinery lives in Data Queries.