WhatsApp Business Agent is in Beta, not yet generally availableBuilt on the Meta Business Agent Platform

Ship AI agents on WhatsApp in an afternoon.

A complete studio for designing, testing, and operating agents on the Meta Business Agent Platform. Skills, knowledge, connectors, evals, live chat, webhooks, all in one workspace.

Self-hostedAccess tokens never leave the serverWorks with any OpenAI-compatible model
$git clone https://github.com/v-senthil/whatsapp-business-agent-studio.git
$npm install && npm run dev
agent-studio · /dashboard/{entityId}/skills
Skills
12 skills · 4 templates available
TemplatesImport CSVNew skill
greeting
Welcome the customer and offer help.
live
order-status
Look up an order by ID and return shipping.
live
handoff-to-human
When frustrated, request a human agent.
live
refund-request
Validate against refund policy, then file.
live
Acme Support
online · agent connected
Hi, order #A-4291 still hasn't shipped.
Checking that now, one moment.
order-status · called
Your package is in transit and scheduled to arrive Friday. Want tracking?
Yes please, and can I change the address?
I can update the address if the order hasn't left the warehouse.
update-address · tool
WhatsApp Cloud APIMeta Agent PlatformClaude Agent SDKOpenAI-compatibleDocker & AppSailNext.js 15

WhatsApp Business Agent is currently in Beta

This feature is not yet released for general availability. To enable it for your WhatsApp Business Account, an admin must open Meta Business Manager and accept the WhatsApp Business Agent terms and conditions.

Enable URL
https://business.facebook.com/latest/whatsapp_manager/business_ai?business_id={Business ID}&asset_id={WABA ID}

Replace {Business ID} and {WABA ID} with your own IDs. If the page loads and prompts you to accept the T&C, your WABA is eligible for the Beta. After signing in, the studio also generates this link for each of your WABAs on the home page.

Everything you need

A studio, not a wrapper.

Every surface of the Meta Business Agent Platform is here, designed to feel native, not like a raw API reference.

Skills, connectors, tools

Build agents like you build software.

Author skills as prompts, wire connectors to your APIs, and expose them as tools the agent can call. Everything is versioned in JSON so you can export a config and reproduce a workspace anywhere.

  • 22 curated skill templates across five categories
  • Draft skills with AI from a plain-English intent
  • 14 connector templates for common SaaS APIs (Shopify, Stripe, HubSpot, Salesforce, Zendesk, Slack, and more)
  • Import a connector plus one tool per operation from any OpenAPI 3.x spec (YAML or JSON)
  • Path / query / header / body params typed end-to-end
  • Every config exportable as versioned JSON

Knowledge base

Upload files, register websites, curate FAQs. Import and export CSV for anything list-shaped.

UI skills

Rich WhatsApp components the agent can send: carousels, CTA buttons, published Flows, images, interactive lists, and location prompts. Each with a trigger instruction and an enable toggle.

Token budgets

Cap token usage over rolling 1, 7, or 14 day windows at the Business Manager level. Multiple caps stack; whichever is reached first takes effect.

Evals + scenarios

Run Meta eval cases with live status polling. Save any test-chat conversation as a scenario, then replay it side-by-side with the current config, per-turn match / differs / error badges.

Live test chat

Talk to your agent inside the studio, with sticky conversation IDs and one-click save-as-scenario.

Webhook receiver

Signed webhook capture with a live SSE feed, filter bar, and full request inspection.

Generate from a document

Drop in a doc, get FAQs, skills, and connector suggestions you can apply with one click.

API playground

Scoped, free-form request builder with quick-call prefills and a JSON viewer for every response.

How it works

From access token to live agent in four steps.

No CLI. No hand-crafted API calls. No copying tokens into curl. Just the flow you'd want.

01

Sign in with your Meta access token

The token never touches the browser. We store it in an encrypted, httpOnly session cookie and inject it server-side on every call.

02

Onboard, or delete, a WhatsApp agent

Pick a WABA, pick a phone, click onboard. The studio provisions the agent through Meta and shows a live six-step checklist (eligibility, business info, first skill, first connector, rollout) so you always know what's next. Need to start over? A guarded Delete agent action removes the agent (and disconnects the integration when it's the last one on the WABA).

03

Configure skills, knowledge, connectors

Author skills, upload knowledge, wire connectors to your existing APIs (14 SaaS templates to start from, or bring your own OpenAPI spec), or import a config JSON you already have.

04

Test, eval, and go live

Chat with the agent in the studio, run eval suites, replay scenarios against your live config, then flip the rollout switch.

Under the hood

Serious tooling for teams that ship.

The studio isn't just a form-and-list app. It's built for the messy realities of running agents in production.

Security by default

  • Access tokens stay server-side. The browser never sees them.
  • Read-only mode locks all non-GET calls to Meta with a single click.
  • Signed webhook verification with timing-safe comparison.
  • All Meta calls go through a single, auditable proxy.

Built for developers

  • Dev drawer with a ring buffer of every request and Copy-as-cURL.
  • Free-form API playground scoped to the current phone.
  • In-app help center with instant search and a per-page Help link.
  • Vendored OpenAPI reference with theme sync.
  • Config export / import as versioned JSON.

Reusable building blocks

  • 22 skill templates across Onboarding, Support, Sales, Escalation, and Utility.
  • UI skill authoring for rich components: carousels, CTA buttons, WhatsApp Flows, images, interactive lists, and location prompts.
  • 14 prefilled connector templates for common SaaS APIs (Shopify, Stripe, HubSpot, Salesforce, Zendesk, Slack, and more), with proper auth scaffolds.
  • Or bring your own: import a connector plus every operation as a tool from any OpenAPI 3.x YAML / JSON spec.
  • CSV import / export for skills, FAQs, websites, allowlist.
  • Save any chat as a scenario; replay it side-by-side against the current config with per-turn match / differs badges.

Live operations

  • Server-Sent Events feed of incoming webhooks with signed/unverified badges.
  • Connector health dashboards with p95 latency and top failure patterns.
  • Live agent event polling: trigger, watch, react.
  • Thread control for graceful human handoff.
  • Token budgets with rolling 1, 7, or 14 day caps to keep spend predictable.

Built-in AI assist

  • Draft skills from a plain-English intent on the new-skill page.
  • Drop in a doc, get FAQs, skills, and connector suggestions with one-click apply.
  • Pluggable providers: local Claude Agent SDK or any OpenAI-compatible endpoint (Ollama, LM Studio, Together, and more).
  • AI keys live server-side; the browser only ever sees whether one is configured.
For product teams

Ship agent behavior without babysitting engineers for every copy tweak.

For engineers

Type-safe schemas, a real API playground, and every call recorded for cURL.

For agencies

Export a client's whole configuration as JSON, replicate it on the next account.

FAQ

Straight answers.

If you have another question, browse the full help center. Every page in the studio also has a Help link that opens the matching article.

What do I need to get started?

A Meta Business account with the Agent Platform enabled, plus a valid access token. You paste the token on sign-in and the studio handles the rest, discovering your WABAs, phone numbers, and existing agents.

Where does my access token live?

In an encrypted, httpOnly session cookie, server-side only. The browser never sees the raw token, and every Meta API call is proxied through same-origin routes that inject it at request time.

Which AI model does the studio use?

The agent itself runs on Meta's Business Agent Platform. Optional assistive features (skill drafting, doc-to-agent generation) call your configured provider: either the local Claude Agent SDK or any OpenAI-compatible endpoint (OpenAI, Ollama, LM Studio, Together, and more).

Can I import a configuration I already built?

Yes. Every workspace can be exported as a versioned JSON bundle covering settings, business info, skills, knowledge, connectors, and tools. Import validates the shape, previews the counts, then applies mutations sequentially with a live progress bar.

Is there a hosted version?

The studio is self-hosted. Ship it as a Docker container (Dockerfile plus docker-compose.yaml included), on Apple's native container CLI, or as a one-shot Zoho Catalyst AppSail deploy. You keep control of the box, the cookie secret, and the access tokens.

How do I find help while I use the app?

Every page has a Help link in its top-right that opens the matching article in a new tab. The full help center at /help has a search box (press / to focus) that ranks matches across article titles, sections, and body text.

How do I test changes safely before going live?

Use the built-in test chat, save any promising conversation as a scenario, and replay it side-by-side against the current config. Each turn is marked match, differs, or error so you can spot regressions at a glance. Combine with the Meta eval suite for structured runs.

Your customers already message you on WhatsApp.

Give them an agent that actually helps. Sign in with a Meta access token and be testing your first agent within the hour.