Wix Base44 Connector

Building on Wix from Base44

The Wix connector is connected; this is how the app gets built on it — gather the site’s context, find the APIs and learn their contracts from the docs, write the code. Discover everything: endpoints, paths, doc URLs, request and response fields all come from the calls below, never from memory or pattern. 404 or empty ⇒ discover, not permute. Examples teach mechanics and go stale — verify before relying.

Start every Wix task by getting the site context (wx.context, below). It reports what the site actually has and how each part is configured, so every call targets what is really there instead of a guess from a name. It is the first step of a task, not a one-time setup: a later turn does not inherit an earlier turn’s context, so re-read it before building on or calling a Wix API.

A management or admin task starts at the recipes: call wx.mgmtRecipes (Learn Wix) and follow the one that fits. A recipe states outright what an API does not support, which fields a bulk call actually writes, and the order two calls have to go in — from the schemas alone those get re-derived several errors at a time. wx.search ranks the matching recipes too (recipe entries in hits), so they also surface from a search that began at the methods.

In this skill:

• What are you building? — for visitors, for the owner, or management done for them — decided per request from the user’s intent, each on its own flow
• The helpers — the wx.* loader every exec opens with
• Gather context — wx.context, the report of what the site actually has
• Learn Wix — find the APIs, learn their contracts
  • Management recipes — first stop for an admin task (wx.mgmtRecipes)
  • Search and browse — wx.search across methods, recipes and articles
  • Read a doc page — wx.page for a located doc URL
  • The spec index — wx.spec, a located method’s exact schema
• Write the code
  • Admin ops while building — you, in exec_tool
  • Backend functions — the app, as the owner
  • Visitor authentication and Wix-hosted flows — the visitor client, redirect sessions, the headless OAuth app — and bootstrapping a whole vertical with wix-vibe-headless

What are you building?

Wix has two identities, visitor and admin, and each has its own flow. Decide which one every request is for before acting on it — from what the user wants, not from the technology:

• Visitor — the people using the site: browsing, booking, buying, signing in as a member. Visitor token, called from the browser through one shared visitor client.
• Admin, built — a management surface of the owner’s own: a custom view, a bulk operation, an internal tool. Admin connector token, server-side in backend functions the frontend calls.
• Admin, done for them — the user wants the business managed now, not an app: create the products, change a setting, import the contacts. Admin token, ad hoc in exec_tool; nothing ships in the app.

The owner already has a full Wix dashboard for running the business — orders, products, bookings, settings, and more — so an admin build is a choice the user makes, not a gap to fill by default. When a request has both visitor and admin parts, each keeps its own flow.

When the request doesn’t say which, ask the user — in their terms (who the thing is for), not in technical ones (identities, tokens, flows). When you decide without asking, tell them what you chose the same way.

The rules that follow from the split:

• Checkout runs on a visitor token, never the admin token — including a checkout this app creates for the visitor. The admin token will hand back a checkout URL, and that is the trap: it belongs to the site rather than the buyer (no cart persistence, no abandoned-checkout recovery, no attribution), and a redirect session refuses it outright (403). Returns reach this app only from URLs in the OAuth app’s redirect list (Visitor authentication and Wix-hosted flows, below).
• No OAuth app yet is not a reason to use the admin token. wx.ensureOAuthApp returns the clientId, creating the app when the site has none — a visitor flow’s first step. Anyone can mint an anonymous visitor token from that public clientId; no visitor login is required.
• No login still means visitors. asServiceRole governs this app’s own data access and says nothing about which identity Wix sees: a public page’s Wix calls run on a visitor token. Reach for the admin token because a feature is owner-side, never because nobody signs in.
• Owner-side work beyond pages — webhooks, scheduled jobs, explicitly authorized elevated operations — is admin flow too, in backend functions.

visitor pages ──(visitor token)────────────────────────► wixapis.com
admin pages   ──► base44/functions/… ──(admin token)────► wixapis.com
exec_tool     ──(admin token, ad hoc management)────────► wixapis.com

The connected site owns its payments. Whatever the visitor buys is paid on the Wix-hosted page this app hands them off to, with the provider the site already has. Base44’s <available_payment_providers> note is about checkout that Base44 hosts instead — it does not apply to these flows, and asks nothing of them, whichever providers it names or rules out.

The helpers

Research and probing run in exec_tool, and one loader opens every exec (execs share no state — reload each round; the module lives on disk next to this file, network only as first-touch fallback):

const fs = require("fs"), P = ".agents/skills/wix-base44-connector/utils.cjs";
if (!fs.existsSync(P)) { fs.mkdirSync(".agents/skills/wix-base44-connector", { recursive: true });
  fs.writeFileSync(P, await (await fetch("https://www.wix.com/skills/wix-base44-connector/scripts/utils.cjs")).text()); }
const wx = require(require("path").resolve(P));

wx exports these helpers:

• wx.post/get/patch/put/del(url, [body], token?) — JSON transports, one per verb (get/del take no body): Bearer from token, non-2xx throws the API’s own error
• wx.clip(value) — cap a return value: oversized → { truncated, total, head }; renders undefined as null so absence stays visible
• wx.context(token) — the site’s full dynamic context report; inline when small, otherwise a saved Markdown file with a heading outline
• wx.browse(menuUrl, { include, filter, depth }) — walk a docs-portal menu deterministically
• wx.search(term, { type, max, lines }) — ranked REST docs, Headless articles, and management recipes in one hits list; see the response type under Search and browse. Only method hits carry an endpoint.
• wx.page(docsUrl) — read a doc page; its worked examples come back as titles + line numbers
• wx.bash(cmd) — shell over saved files (GNU grep/sed; awk is mawk; no rg)
• wx.spec(docsUrl | code) — a method’s exact schema, plus the titles of the docs’ own request examples saved at examplesPath; pass a hit’s docsUrl (direct load), or raw code to query the index yourself
• wx.mgmtRecipes(q?) — management-recipe index; no arg → categories, a word → matching recipes
• wx.ensureOAuthApp(token, { name, redirectUris, redirectDomains }) — the headless OAuth app, found by name or created, with the redirect lists merged into whatever it already allows; returns { clientId, created }. clientId is what the browser mints visitor tokens from. Run it whenever context reports no OAuth app, or before any Wix-hosted flow that returns to this app — see Visitor authentication and Wix-hosted flows
• wx.installApp(appDefId, siteId, token) — install a Wix app on the site (Apps Installer). If discovery finds an API whose app isn’t installed on the site, install it first — that’s a one-call prerequisite, not a reason to fall back to a hand-built alternative. appDefId from search or the Apps-Created-by-Wix table; siteId from context (the site report)

Search makes one combined request across REST, management recipes, and Headless by default. The service ranks them together and interleaves methods and articles; the helper preserves that order. max limits the combined result count (default 15).

Every helper answers inline when the result fits (≤ 4,000 chars — exec results clip at ~5,000). A bigger result is saved under .agents/skills/wix-base44-connector/tmp/ and comes back as { path, bytes, lines, outline } — the outline is your map into the file. Work a saved file in two moves: find with wx.bash("grep -n 'term' <path> | head -40") (or across every save: grep -rn 'term' .agents/skills/wix-base44-connector/tmp/), then quote with read_file — an offset/limit window at the lines grep named, or the whole file when it fits the 45K cap.

Gather context — the dynamic context report

const { accessToken } = await base44.asServiceRole.connectors.getConnection("wix");
return await wx.context(accessToken);

One report: installed apps with ids (incl. Stores’ catalog version — V1 vs V3 decides its endpoints), the OAuth app id (also the visitor clientId), locale, currency, CMS collections. An empty report = bad token, never an empty site.

No OAuth app in the report means the site has none yet. For anything a visitor operates, create it with wx.ensureOAuthApp — one call, returns the clientId the frontend mints visitor tokens from — and build the feature on the visitor flow.

Reports over 4,000 characters are saved in full to a temporary Markdown file. The result includes its path, byte and line counts, and a heading outline. Read that file to inspect the site context.

Learn Wix — find the APIs, learn their contracts

Management recipes — first stop for an admin task

Curated admin flows, by category — ecommerce, bookings, stores, cms, google-ads, sites, contacts, get-paid, marketing, pricing-plans, events, blog, forms, restaurants, domains, media, …:

await wx.mgmtRecipes();           // categories with counts
await wx.mgmtRecipes("stores");   // a category's list — or any task word: wx.mgmtRecipes("coupon")
// each row points at its recipe: `file` when the wix-manage skill is installed — read_file it
// straight off disk — else `url`: wx.page(url), whole when small, saved + outline when big

A recipe carries prerequisites, order, and gotchas that no method page has. wx.search ranks these same recipes in its hits list, so they surface either way.

Search and browse

wx.search returns a saved-file reference and ranked hits. Each hit is a method, article, or recipe, identified by its title field; only methods have an endpoint.

type Section = { title: string; line: number }; // 1-based line in the saved file
type SearchResult = {
  path: string; bytes: number; lines: number; note?: string;
  hits: ({ docsUrl: string } & (
    | { method: string; endpoint: string | null; gist?: string; examples?: Section[] }
    | { article: string | null; line: number; outline?: Section[] }
    | { recipe: string | null; line: number; lines: number;
        file?: string; steps?: string[]; calls?: string[] }
  ))[];
};

The file at path keeps the full returned Markdown, while optional details and excess hits may be omitted from the inline response. Null titles or endpoints were not parsed. If no result blocks parse, head and note replace hits; the final size-limit fallback is { truncated: true, total: number, head: string } instead of the object above.

// name the method? search finds it in one call:
const result = await wx.search("stores v3 update product"); // SearchResult above
// Keep all hit kinds; only a method hit has an endpoint to call and a schema for wx.spec.
// exploring an unfamiliar product? browse is deterministic — menuUrl alone orients (children + counts);
// filter before listing methods. browse works for both portals this skill uses — REST
// (api-reference) and WIX_HEADLESS (go-headless) — just pass that portal's menu URL.
await wx.browse("https://dev.wix.com/docs/api-reference/business-solutions/bookings/bookings",
                { include: ["METHOD"], filter: "resched", depth: 4 });
// non-REST portal — same call, that portal's menu URL:
await wx.browse("https://dev.wix.com/docs/go-headless/authentication", { depth: 2 });

// don't know where it lives? search ranks, never says "no match" — drop wrong-product hits
await wx.search("pause a pricing plan subscription and resume it");
// Same SearchResult: methods, articles, and recipes remain in ranked order.

Products and their capabilities — the common ones, partial lists:

• Stores — products · categories · product options and variants · inventory · promotions · +7 more
• Bookings — services · appointments · classes · staff members · time slots · waitlists · +7 more
• eCommerce — cart · checkout · orders · order fulfillment · discount rules · +7 more
• Events — events · ticket definitions · RSVP · check-in · +6 more
• Restaurants — menus · items · item modifiers · online orders · reservations · +9 more
• Blog — posts · draft posts · categories · tags · +3 more
• CMS — data items · data collections · collection permissions · external databases · +4 more
• Pricing Plans — plans · orders · recurring subscriptions · free trial periods · +4 more
• Members & Contacts — contacts · labels · extended fields · members · badges · +7 more
• Forms — form schemas · form submissions · interactive form sessions · +3 more
• Loyalty — loyalty points · earning rules · tiers · rewards · +5 more

Full list — all 36 products, their capabilities and docs paths: references/CAPABILITY_MAP.md.

Go deeper for fields, enums, or absence — only the spec index proves absence.

Read a doc page

Method pages are 100 KB+, twin REST and SDK halves repeating field names at different types — map and window in the SAME exec; coordinates are for your code, not for a second round:

const pg = await wx.page(docsUrl);   // whole text when small; else { path, bytes, lines, outline }
// pg.examples lists the page's worked requests as { title, line } — a complete request (URL,
// headers, body) is usually all you need: read_file(pg.path, offset: <its line>)
// anything else, windowed to your term in the same round:
return wx.bash(`sed -n '/^## REST API/,/^## JavaScript SDK/p' ${pg.path} | grep -B5 -A40 -i 'examples\\|<term>' | head -c 3800`);
// a giant fenced example → its field vocabulary instead of paging it:
// wx.bash(`sed -n '<a>,<b>p' ${pg.path} | grep -oE '"[a-zA-Z]+":' | sort -u | head -60`)

search also saves its raw content beside the inline hits — grep its path when a hit’s six lines weren’t enough.

The spec index — a located method’s exact schema

Pass a method’s docsUrl (a search/browse hit carries it) — spec loads that method’s schema (request body, responses, filterable-fields map, examples) in one call, a direct lookup. Read the field descriptions, not just the names — they carry the rules (which field is canonical, when one is empty):

const sp = await wx.spec(hit.docsUrl);
// sp.examples — the method's worked requests as { title, line }, all saved at sp.examplesPath.
// The one that matches the task is rarely the first: read_file(sp.examplesPath, offset: <its line>)

Want to shape the result yourself? Pass raw code against the index (lightIndex + getResourceSchemaByUrl):

await wx.spec(`
  const url = "<docsUrl from search/browse>";           // API method page, not a skill/article page
  const s = await getResourceSchemaByUrl(url);
  // A method URL resolves to exactly that method — take methods[0]. Don't match on docsUrl:
  // the reader normalizes the url, so equality against the one you passed can miss.
  const m = s.methods[0];
  return {
    call: m.publicUrl,                                       // callable https://www.wixapis.com/… URL
    body: m.requestBody?.content["application/json"].schema.properties,
    responses: m.responses,
    filterable: m.queryMethodData?.queryFieldsCapabilitiesMap      // query methods
             || m.searchMethodData?.searchFieldsCapabilitiesMap,   // search methods
    example: m.legacyExamples?.[0]?.content,
  };
  // { $circular: "<name>" } types resolve via s.components.schemas["<name>"] — complete in this call
`);

filterable maps each field to its allowed operators + sort — filter server-side only on what it lists, else filter in code.

Write the code

REST, not the JS SDK — everywhere: the admin ops you run while building, backend functions, frontend pages. Doc pages carry twin REST and SDK halves; read the REST one.

Send what the page documents, nothing more — required fields, nesting, and enum values exactly as spelled; a header only when the operation’s REST section lists it. Never wix-site-id: every token here is already bound to a site, and in the browser the header fails CORS preflight outright.

Shapes are discovery too: read a method’s request/response schema from spec() — its field descriptions state which field is canonical and when one reads back empty. Don’t infer shape from a single live probe: it reflects only the params you sent, so probe with the same request your code makes.

Admin ops while building — you, in exec_tool

const { accessToken } = await base44.asServiceRole.connectors.getConnection("wix");
const data = await wx.post("https://www.wixapis.com/contacts/v5/contacts/query",   // spec-index publicUrl
  { query: { cursorPaging: { limit: 10 } } }, accessToken);
// let it throw — the thrown message is your result; .catch hides the answer
return wx.clip({ error: null, count: data.contacts?.length, first: data.contacts[0] });
// `first` is one complete record — read the real field shapes off it; don't code from remembered key names

Backend functions — the app, as the owner

The same API call, deployed. No helpers run here — wx.* is a build tool loaded into exec, absent from base44/functions/…; write plain fetch. Nor clip, which caps what an exec returns to you: a function returns its data to the app.

// inside base44/functions/… — plain fetch, no wx
const { accessToken } = await base44.asServiceRole.connectors.getConnection("wix");
const res = await fetch("https://www.wixapis.com/contacts/v5/contacts/query", {
  method: "POST",
  headers: { Authorization: `Bearer ${accessToken}`, "Content-Type": "application/json" },
  body: JSON.stringify({ query: { cursorPaging: { limit: 10 } } }),
});
if (!res.ok) throw new Error(`${res.status} ${await res.text()}`);   // the API's own error, not a generic one
return (await res.json()).contacts;

• That Authorization: Bearer … is the whole auth for this lane, unless the operation’s page says otherwise.
• One file per business area, not per call — each file is its own deploy, and deploys cost time.
• Call every function you deploy and fix what breaks. Deploying is not testing.

Visitor authentication and Wix-hosted flows

The “site for visitors” shape (What are you building?), in code — one file pages import. Neither clientId (from the context report) nor the minted token is a secret; together they are “an anonymous visitor”, safe in shipped code:

// src/lib/wixClient.js
let token;
const mint = async (body) => {
  const r = await (await fetch("https://www.wixapis.com/oauth2/token", { method: "POST",
    headers: { "Content-Type": "application/json" }, body: JSON.stringify(body) })).json();
  token = r.access_token; sessionStorage.setItem("wixRefresh", r.refresh_token);   // expires_in: 14400s = 4h
};
// first visit:  mint({ clientId: WIX_CLIENT_ID, grantType: "anonymous" });
// on expiry:    mint({ refreshToken: sessionStorage.getItem("wixRefresh"), grantType: "refresh_token" });
export const wix = (path, opts = {}) => fetch("https://www.wixapis.com" + path, { ...opts,
  headers: { Authorization: `Bearer ${token}`, "Content-Type": "application/json" } });

If you need to test a visitor API, including redirect sessions, you can do so in exec_tool. Mint a visitor token and pass it to the API call so the test uses the same identity as the visitor frontend. Token contract: Retrieve Tokens (opens in a new tab). For example, test a public read:

const { access_token: visitorToken } = await wx.post("https://www.wixapis.com/oauth2/token",
  { clientId: WIX_CLIENT_ID, grantType: "anonymous" });   // clientId: from the context report
return await wx.post("<a public read from Learn Wix>", { query: {} }, visitorToken);
// 200 ⇒ every visitor-facing page in the app is this same call, no server between
// a lean default response isn't the whole shape — contracts often define a fields param
// that opts INTO heavier parts (formatted prices, media); read the contract for it

Any Wix-hosted flow that returns the visitor to your app needs the headless OAuth app’s redirect config set. This covers redirect sessions and sending a buyer to the Wix-hosted checkout checkoutUrl and back — any flow where Wix redirects to a URL on your app.

Step one of any visitor flow: wx.ensureOAuthApp with this app’s own URLs. It returns the clientId the frontend mints visitor tokens from (anonymous visitors do not need to log in), creating the OAuth app when the site has none and adding your URLs to its lists when it already has one. Returns only complete for URLs in those lists, and a missing one breaks silently — create-checkout and the anonymous token still succeed — until a real buyer is redirected and stranded on Wix. It is idempotent by name: never create a second OAuth app for the same Base44 app, or the first clientId‘s returns stop working.

// Register BOTH of this app's own URLs — its preview URL and its published URL — so returns work
// before AND after publish. Both are built from the Base44 app id (NOT the Wix OAuth client_id /
// appId used above):
//   preview:   https://preview-sandbox--<base44-app-id>.base44.app
//   published: https://<app-name-slug>-<last 8 of the base44-app-id>.base44.app
//              where <app-name-slug> is the app's name (the brand name) slugified — e.g.
//              "Flash Sale Spark" -> flash-sale-spark-a4615088.base44.app.
// A custom domain the owner connects overrides the published one; the app name (hence the slug)
// can also change. So prefer the app's real current published URL when you have it, and update
// this same OAuth app whenever the name or domain changes — never create another.
const appOrigins = [previewUrl, publishedUrl]; // this app's actual preview + published URLs
const loginCallbacks = appOrigins.map(origin => new URL("/login-callback", origin).href);
const returnDomains = appOrigins.map(origin => new URL(origin).hostname);

// OAuth redirect configuration: exact login URLs versus domains for other returns.
// https://dev.wix.com/docs/go-headless/authentication/setup/allow-redirect-uris-and-domains.md
const { accessToken: adminToken } = await base44.asServiceRole.connectors.getConnection("wix");
const { clientId, created } = await wx.ensureOAuthApp(adminToken, {
  name: "My App",                     // this app's name; the lookup key on later runs
  redirectUris: loginCallbacks,       // login callbacks: matched exactly by the auth request
  redirectDomains: returnDomains,     // returns from Wix-hosted flows: hostnames only
});
// clientId is the public visitor client id the frontend client above mints from.
// When destinations change later — a custom domain, a renamed app — call this again with the
// new URLs: it updates the same OAuth app and preserves the entries already there.

Frontend redirect example, using the visitor client above after obtaining a visitor token:

import { wix } from "@/lib/wixClient";

// Flow prerequisites and supported intents:
// https://dev.wix.com/docs/go-headless/business-solutions/wix-hosted-pages/redirect-using-the-rest-api.md
export async function redirectToWix(intent, returnPath = "/") {
  // wix sends the minted visitor token. Never use the admin token here.
  // Admin tokens are for ad hoc management calls or backend functions implementing admin logic.
  // Pass the intent required by the selected flow's schema.
  const response = await wix("/headless/v1/redirect-session", {
    method: "POST",
    body: JSON.stringify({
      ...intent,
      callbacks: {
        postFlowUrl: new URL(returnPath, window.location.origin).href,
      },
    }),
  });
  if (!response.ok) throw new Error(`${response.status} ${await response.text()}`);
  const { redirectSession } = await response.json();
  window.location.assign(redirectSession.fullUrl);
}

Building a whole vertical? Bootstrap it with wix-vibe-headless. When the build is a visitor-facing site on a Wix vertical it supports — a storefront, bookings, pricing plans, events, restaurants, rentals, a blog and more — the wix-vibe-headless skill stands the vertical up fast: it is built entirely on the same Wix APIs as this skill, but ships working code (REST transport, scaffolds, UI) plus management-API abstractions (seeding, configuration), so the vertical bootstraps on Base44 in a fraction of the calls. Read the entry point at www.wix.com/skills/wix-vibe-headless/platforms/base44.md (opens in a new tab). Its STEP 1 carries the install itself — the verbatim npx skills add call to run through exec_tool, the companion skills, the vertical table and the scaffold deploy — after which the guide lives at .agents/skills/wix-vibe-headless/platforms/base44.md. Reading it as reference alone is fine, but when you do install, go through its flow end to end — the shipped scaffolds and seed path are most of its value.