> ## Documentation Index
> Fetch the complete documentation index at: https://www.voiceflow.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Connect Voiceflow to Chatwoot

> Build and test a two-way live agent handoff connection with Chatwoot. Customers stay in Voiceflow while your team replies from your support platform.

## What you will build

Customers stay in their Voiceflow conversation while your team replies from Chatwoot. Your service translates the platform’s conversation operations and events into the [live agent handoff system](/docs/documentation/handoff/overview).

This is a developer implementation guide, not a native connector. The API approach has been reviewed against provider documentation; a live account test is required before rollout.

## Before you begin

You need a user API token, account ID, API inbox, and a test contact associated with that inbox, an HTTPS service, and access to Custom Handoff in Voiceflow. Start with [the shared connection starter](/docs/developers/handoff/connect). Choose one test customer and one support destination.

The supplied pilot fixes one test contact and inbox. Replace this with customer lookup and persistent mapping before serving multiple customers.

## Implement the connection

<Steps>
<Step title="Prepare the support workspace">
Create an API inbox and associate a test contact with it. Record the contact’s inbox `source_id`, contact ID, and inbox ID. The source ID identifies the contact-inbox relationship; it is not interchangeable with a user ID.

Confirm a teammate can receive and answer a manually created test conversation before adding Voiceflow.
</Step>
<Step title="Implement conversation creation">
POST `/api/v1/accounts/{account_id}/conversations` with `source_id`, `contact_id`, `inbox_id` and `status: "open"`. Save the returned `id`.

Example provider request body for `POST /api/v1/accounts/ACCOUNT_ID/conversations` (replace illustrative IDs):

```json
{
  "source_id": "CONTACT_INBOX_SOURCE_ID",
  "contact_id": 1,
  "inbox_id": 1,
  "status": "open"
}
```

Use the starter’s `create(request, connectionId)` hook. Persist the provider ID and connection ID together. Forward relevant context as an initial summary or supported private note; never inject internal metadata into customer-visible replies.

The starter includes this text-only pilot. Set `CHATWOOT_URL`, `CHATWOOT_ACCOUNT_ID`, `CHATWOOT_TOKEN`, `CHATWOOT_INBOX_ID`, `CHATWOOT_CONTACT_ID`, and `CHATWOOT_SOURCE_ID`, then run `PROVIDER=chatwoot node server.mjs`. Use the test contact only. Register a Chatwoot account webhook at your service’s `/provider-events` URL for `message_created` and `conversation_status_changed`. Set `CHATWOOT_WEBHOOK_SECRET` to the secret shown by Chatwoot. The pilot validates Chatwoot’s timestamp and raw-body signature, filters public outgoing replies, and maps resolved conversations to closure. [Chatwoot webhook authentication](https://app.chatwoot.com/hc/user-guide/articles/1677693021-how-to-use-webhooks).
</Step>
<Step title="Forward customer messages">
POST `/api/v1/accounts/{account_id}/conversations/{id}/messages` with `content`, `message_type: "incoming"`, and `private: false`.

Implement the starter’s `message(providerConversationId, message)` hook. Deduplicate the Voiceflow message `id`. For the first test, support text and explicitly reject unsupported files rather than acknowledging and dropping them.
</Step>
<Step title="Receive human replies">
Subscribe to message and conversation-status events. Relay only outgoing human messages where `private` is false; exclude incoming messages and activity entries. Map resolved conversations to `conversation.ended`.

The supplied `/provider-events` receiver handles this for the pilot. In another implementation, after provider authentication succeeds, look up the mapped connection ID and send a normalized `message.received` event to the starter’s authenticated `/events` route. Include a stable provider event ID, timestamp and human display name. Do not manufacture an `agent.joined` event merely because a conversation was created.
</Step>
<Step title="Implement closure">
POST `/api/v1/accounts/{account_id}/conversations/{id}/toggle_status` with `status: "resolved"`.

Implement the `end(providerConversationId)` hook. Deliver `conversation.ended` for provider-initiated closure. Make both paths idempotent so a reflected closure cannot trigger an endless loop.
</Step>
<Step title="Connect and test in Voiceflow">
Configure your service in **Settings → Integrations → Custom Handoff**, then follow [Set up live agent handoff](/docs/documentation/handoff/setup). Set Queue ID only when your service has a documented mapping. Publish to a test environment and use the actual customer channel.
</Step>
</Steps>

## Verify a complete handoff

Run this acceptance sequence with a teammate in Chatwoot:

1. Escalate and confirm exactly one conversation appears in the intended inbox.
2. Check the customer identity and escalation context.
3. Send “customer test 1” from Voiceflow and “human test 1” from the support inbox. Both must arrive once.
4. Add a private note; it must not reach the customer.
5. Replay the same customer-message ID and provider event ID; neither should duplicate a message.
6. Close from the support inbox and verify the configured return-to-AI path.
7. Repeat with invalid credentials and an unavailable provider; verify the fallback.

If messages disappear, inspect the provider delivery log, the ID mapping and Voiceflow callback response in that order. A successful create call alone does not establish that the connection works.

## Reference and next steps

[Provider setup/API documentation](https://developers.chatwoot.com/api-reference/conversations/create-new-conversation) · [Provider messages/events documentation](https://developers.chatwoot.com/api-reference/messages/create-new-message)

[Connection API](/docs/developers/handoff/connection-api) · [Event contract](/docs/developers/handoff/webhook-events) · [Troubleshooting](/docs/documentation/handoff/troubleshooting)

Before production, replace in-memory state with durable storage, implement the provider’s retry/refresh rules, and test attachment and routing support only if your deployment needs them.