Menu

Mattering Automations tawk.to integration

Connect tawk.to to Mattering Automations: list properties, read chat history and tickets, read and create Knowledge Base articles and categories, subscribe and unsubscribe webhooks, and call any other method by name, authenticated with one account-wide REST API key.

The Mattering Automations tawk.to integration puts tawk.to REST API v1 on your automation canvas: list the properties an account can see, read chat history and tickets, read and create Knowledge Base articles and categories, subscribe and unsubscribe webhooks, and reach any other method by name. 11 nodes in total, 6 reads and 5 actions.

Read the access section before you plan a project around this. The tawk.to live chat product is free and needs no approval. The REST API is a separate thing and it is not self-serve: a key can only be created after tawk.to grants access in response to a request form on its site, and until then the REST API Keys tab does not appear in your profile at all. Approval is manual, and the public record on how long it takes ranges from tawk.to saying it is typically under a day to users on its own forum reporting no answer after months. Submit the request early.

The credential itself is simple once you have the key: one field, applied as HTTP Basic with the key as the username and an empty password. It is scoped to the ACCOUNT rather than to a property, so one key sees your whole estate and the property is chosen per node from a picker. Setup takes a few minutes, and this guide walks through each step and ends with the full node reference.

Before you start

  • A tawk.to account with at least one property. A property is one chat widget on one site, and its id is required by every node except the agent read and the custom method node.
  • REST API access granted by tawk.to. This is requested through the REST API beta access request form on tawk.to and approved manually. Approval also brings the username, password and link for the reference documentation, which tawk.to serves behind a password rather than publishing.
  • A REST API key created after that approval, under your profile image in the dashboard: Edit Profile, then REST API Keys, then Create Key.
  • Realistic expectations about scope, because four things a reader usually assumes are not available on this API and tawk.to has confirmed each of them. There is no ticket creation, no contact create or sync, no way to fetch live or in-progress chats, and no ticket replies in a ticket read. Real time comes from webhooks, and visitor identity is a job for the JavaScript widget rather than the REST API.
  • A destination for the data that is appropriate for personal data. Chat and ticket payloads carry visitor names, email addresses, cities, countries and full verbatim transcripts including uploaded file links.
  • A Mattering workspace with access to the Automations app.

Request access and create a tawk.to REST API key

tawk.to authenticates every REST call with one API key, sent as HTTP Basic with the key as the username and an empty password. There is nothing else to configure: one global host, one account-wide key, and no per-property secret. The only hard part is getting the key at all, which is a manual approval rather than a form you fill in and use immediately.

Open the REST API beta access request form on tawk.to and submit a request under the identity you intend to run automations as. Approval is manual and is not immediate.
Do not build a delivery date around this step. tawk.to has said approval typically takes under a day, and there are users on its own forum who applied and never received a response, one describing a wait of nearly a year. Request access at the start of a project, not at the end.
When the approval email arrives, sign in to the tawk.to dashboard and click your profile image at the lower left.
Choose Edit Profile, open the REST API Keys tab, and click Create Key. Copy the key.
If the REST API Keys tab is not there, the account has not been granted access yet. Its absence is what people usually report as not being able to find the screen, and no setting inside the dashboard reveals it.
The key is per ACCOUNT, not per property. It can see every property the account reaches, which is why the property is chosen on each node rather than stored here, and why this credential has a single field.

Connect tawk.to to Mattering

In Mattering Automations, open Connectors and click Add Connector.
Choose tawk.to from the list.
Click Create new credential, paste the REST API key, and save.
There is deliberately no password field. tawk.to has no password here: the header is the key as the username with an empty password, and a client that supplies one gets a 401 on every method that reads like a bad key rather than a bad form.
Click Test Connection. Mattering posts to the agent endpoint with an empty JSON body, which is tawk.to's own worked example and the only method that needs no property id.
A pass proves the key is valid, that the Basic construction is right and that the empty-body convention is being honoured. It does not prove any particular property is reachable, and it does not prove the Knowledge Base methods will work, because those carry server-side faults of their own.
Click Connect. All 11 tawk.to nodes appear in your canvas palette.

Verify it works

Open any automation canvas and drop tawk.to: List Properties onto it, then run a test step with no configuration at all. It returns every property the key can see, and it is the root of this integration's picker chain, so a successful run here is also what makes the property field on the other eight nodes populate. Note the response shape while you are here: tawk.to wraps its payload in its own envelope, so paths read data.data[].propertyId and data.data[].name, with two levels of data rather than one. This is the single most common thing to get wrong when hand-writing a path.

Now confirm the picker. Drop tawk.to: List Chats onto the canvas and open its Property field: it should list your properties by name rather than being a free-text box. Choose one, set a narrow date range covering a day you know had traffic, and run the step. You should get chat rows back, capped at 50 per call. Treat what comes back as personal data from the first run: it carries visitor names, email addresses and full transcripts, and it lands in the run log of whichever workspace ran it.

If your account has a Knowledge Base, tawk.to: List Knowledge Base Articles is a useful third check, because the Knowledge Base methods are the ones with known server-side faults and a green connection test says nothing about them. Everything in this verify path is read-only, so none of it can reach tawk.to: Delete Webhook, and it is safe to re-run whenever you rotate the key.

Node reference

11 nodes, generated from the connector itself so this reference always matches what ships.

Read and query nodes 6

tawk.to: Get Current Agent

tawk_to_get_current_agent

Return the agent the REST API key belongs to. This is the vendor's own worked example and the cheapest way to prove a key works, so it is this connector's auth probe. It takes no parameters, but tawk.to still requires a JSON body, so the node always sends an empty object; a request with no body at all fails with validation_error on 'body'. A POST despite being a pure read: every tawk.to method is a POST.

tawk.to: Get Knowledge Base Article

tawk_to_get_kb_article

Fetch one Knowledge Base article by id. Two quirks are worth knowing before you debug this node: tawk.to requires the category id here as well as the article id, and it rejects a real Knowledge Base site id, accepting only 'primary'. Both are the vendor's behaviour, reproduced faithfully. Asking for the content block has also been reported to return a server error while the meta block succeeds, so a 500 here is a known gap rather than a configuration mistake.

tawk.to: List Chats

tawk_to_list_chats

List past chats for a property, optionally narrowed to a date range. This is chat HISTORY, not a live feed: tawk.to confirms in-progress chats are not returned, so use a webhook if you need real time. A call returns at most 50 chats and there is no cursor, so walk longer histories by moving the dates. Chat payloads carry visitor names, email addresses and full message text, so treat the output as personal data. tawk.to publishes no cursor and no page token: walk history by moving the start and end dates, not by paging.

tawk.to: List Knowledge Base Articles

tawk_to_list_kb_articles

List the Knowledge Base articles on a property, optionally narrowed to one category. Use it to find the article id that Get Knowledge Base Article needs, or to audit which articles are still drafts. Unlike the single-article read, this method accepts a real Knowledge Base site id as well as 'primary'.

tawk.to: List Properties

tawk_to_list_properties

List every property the API key's account can see. A property is one chat widget on one site, and its id is required by every other tawk.to node, which makes this the root of the connector's picker chain. Because the key is account-scoped rather than property-scoped, this returns the whole estate. Takes no required parameters; the node sends an empty JSON body when none are set, which tawk.to requires.

tawk.to: List Tickets

tawk_to_list_tickets

List tickets for a property, filterable by status, date range and tags. Each ticket carries its opening message only: tawk.to confirms follow-up replies are not returned by this method and has no committed timeline for adding them. Ticket creation is not available on the REST API either, so this is a read-out, not a round trip. tawk.to publishes no cursor and no page token: walk history by moving the start and end dates, not by paging.

Action nodes 5

tawk.to: Create Knowledge Base Article

tawk_to_create_kb_article

Create a Knowledge Base article on a property. The node assembles tawk.to's three-deep payload for you from flat fields, and deliberately omits the category NAME that the vendor's documentation still shows: sending it is rejected, and tawk.to has confirmed the docs are wrong there. Defaults to a draft, so a test run cannot publish to real readers - which matters because the REST API publishes no article DELETE, so removing an article means opening the dashboard. The slug must be unique across the Knowledge Base or the call fails with a server error rather than a helpful message.

tawk.to: Create Knowledge Base Category

tawk_to_create_kb_category

Create a Knowledge Base category on a property, and return its id - which is worth having, because tawk.to publishes no category lister, so this node and the dashboard URL are the two ways to learn a category id. The node assembles the nested payload from flat fields. There is no category delete on the REST API, so cleaning up a test category means using the dashboard.

tawk.to: Create Webhook

tawk_to_create_webhook

Subscribe a URL to tawk.to events on a property, and return the hook id. This is how you get real time out of tawk.to: chats cannot be polled live, so a webhook is the only way to react to one while it is happening. tawk.to signs every delivery with HMAC-SHA1 in the X-Tawk-Signature header using the property's webhook secret, retries a failed delivery for up to 12 hours, and treats no response within 30 seconds as a failure. Keep the returned hook id: Delete Webhook needs it and there is no lister to recover it from.

tawk.to: Custom API Method (any method)

tawk_to_request

Call any tawk.to REST method by name. tawk.to publishes its REST reference only to accounts approved through a manual beta form, so the typed nodes here cover the ten methods evidenced in public sources and this node covers everything else for anyone holding that access. Every tawk.to method is a POST and requires a JSON body, so send at least an empty object.

tawk.to: Delete Webhook

tawk_to_delete_webhook

DESTRUCTIVE. Remove a webhook subscription from a property. There is no restore and no undo: the hook id is gone, a replacement gets a new one, and anything keyed to the old id breaks. tawk.to also publishes no webhook LISTER on the REST API, so a hook id you did not record when you created it can only be found in the dashboard. Deliveries stop immediately, which means events that happen after this call are lost rather than queued.

Using the integration

On the automation canvas

tawk.to nodes chain with every other node. A weekly schedule can drive tawk.to: List Tickets into a transform and then into a Mattering database, tawk.to: List Chats can feed a summary into a Mattering Comms thread so the team reads a digest instead of a dashboard, and tawk.to: Create Knowledge Base Article can turn a recurring question into a drafted answer. One value does most of the wiring: the property id from tawk.to: List Properties, which every node except tawk.to: Get Current Agent and tawk.to: Custom API Method (any method) needs, and which is a picker rather than a field you type.

From AI assistants over MCP

Every node speaks MCP, the open protocol for connecting AI assistants to tools, so Claude, ChatGPT, and Matter, the assistant built into the platform, can drive the same surface. tawk.to publishes no MCP server of its own, so there is no first-party tool to defer to here. The comfortable surface to hand an assistant is the six reads plus the two Knowledge Base creates, which default to drafts. Keep tawk.to: Delete Webhook out of an agent's toolset: it is the one destructive node here, and a hook id it removes cannot be recovered from any lister. One caution that matters more on this connector than on most: select the fields a step needs rather than handing a whole chat object to an assistant, because a chat object is somebody's transcript.

Why the property is chosen on the node

The key is minted on a profile screen rather than a property screen, and it can see every property the account reaches. One credential therefore spans many properties, so the property belongs on the node rather than on the credential. The field is a picker sourced from tawk.to: List Properties, showing each property by its name and saving its id. Two practical consequences. First, an account that has accumulated sites over the years will show more properties than the person building the workflow expects, so check the value on anything that writes. Second, there is nothing to reconfigure when a new site is added: it appears in the picker as soon as it exists.

Reading history when there is no pagination

There is no pagination anywhere in this API. tawk.to publishes no cursor, no page token and no offset, and no response indicates whether more rows exist. A chat list returns at most 50 results per call. The total on a list response is a count of matched records rather than a page count. So the window IS the pagination: set Start Date and End Date on tawk.to: List Chats or tawk.to: List Tickets, keep the window small enough that a result never comes back at the cap, and walk forward. That arithmetic belongs in the automation, where it is visible, rather than hidden in the connector. tawk.to also publishes no rate limit and no retry contract, so treat throughput as unknown rather than generous, pace bulk reads, and put any backoff in the workflow.

Building Knowledge Base content, and the one-way door

tawk.to: Create Knowledge Base Article assembles tawk.to's three-deep payload from flat fields, and it deliberately omits the category name that the vendor's documentation still shows, because sending it is rejected and tawk.to has confirmed the documentation is wrong there. Two rules make the difference between a smooth run and a confusing one. The slug must be unique across the Knowledge Base, or the call fails with a bare server error rather than a message naming the slug, which is why the slug is a required field rather than something derived from the title. And the node defaults to a draft on purpose: the REST API publishes no article delete and no category delete, so anything created here can only be removed by hand in the dashboard. tawk.to: Create Knowledge Base Category returns a category id worth capturing, since tawk.to publishes no category lister and that response is one of only two places the id appears.

Real time through webhooks

In-progress chats cannot be read, so anything that needs to react while a conversation is happening runs on webhooks. tawk.to: Create Webhook subscribes a URL to a property's events and returns a hook id; point it at a Mattering inbound webhook trigger and a workflow runs on chat start, chat end, transcript created or ticket created. tawk.to signs every delivery with HMAC-SHA1 in a signature header using the property's webhook secret, retries a failed delivery for up to 12 hours, and treats no response within 30 seconds as a failure. Store the hook id the moment you get it: tawk.to publishes no webhook lister, so tawk.to: Delete Webhook can only be pointed at an id you kept or one read out of the dashboard.

Reaching methods the typed nodes do not cover

tawk.to: Custom API Method (any method) calls any tawk.to method by name with a JSON body you supply, and on this connector it is not a footnote. Because the REST reference is served behind a password, the typed nodes cover the ten methods that public sources evidence rather than everything the API has, and the known method names use three different conventions between them, so the rest genuinely cannot be guessed. tawk.to advertises capabilities whose method names are nowhere public, including creating properties, inviting team members, listing property members, chat statistics by agent and tag, and the list and update halves of webhook management. If your approved documentation names one, call it here: the Basic auth, the mandatory JSON body and the error handling are already correct.

Treating the output as personal data

This is unusual enough to state rather than assume. A chat payload carries the visitor's name, email address, city and country plus the complete verbatim transcript including links to uploaded files. A ticket payload carries the requester's name and email and the full body of their message. Three habits keep that under control. Select the fields a downstream step needs rather than forwarding a whole chat or ticket object into a spreadsheet, a chat message or an AI step's context. Be deliberate about which systems a transcript is allowed to reach, since a workflow makes that copy silently. And when testing, prefer a property with low or synthetic traffic, because a test run writes real visitor data into the run log of whichever workspace ran it.

Troubleshooting

There is no REST API Keys tab in my profile
The account has not been granted REST API access yet. The tab is hidden until tawk.to approves an access request, and no setting inside the dashboard reveals it, so this is not a permissions toggle you have missed. Submit the REST API beta access request form on tawk.to and wait for the approval email, which also carries the username, password and link for the reference documentation. Approval is manual and the reported wait varies widely, so treat the key as arriving on a timeline you do not control and build the rest of the workflow against the palette in the meantime.
Every call fails with an authentication error even though the key looks right
Three shapes of the same header fail identically here. The key goes in as the USERNAME with an EMPTY password, so the credential is the key followed by a colon and nothing after it; a client that supplies a password, or one that drops the trailing colon, gets a 401 that reads like a bad key. Mattering builds that header for you and this credential has a single field with no password input for exactly this reason, so if you are seeing this from a Mattering node, check for whitespace pasted around the key and then re-run Test Connection. If you are seeing it from a hand-rolled HTTP node instead, that is the likely cause.
A request fails with a validation error naming the body
That is tawk.to rejecting a request that carried no JSON body at all, and it is the single most common failure on hand-built tawk.to calls. Every method here is a POST and every one of them requires a body, even the methods that take no parameters, which must be sent an empty object. The tawk.to nodes never send an empty request, including tawk.to: Custom API Method (any method), so this points at a generic HTTP node rather than at this integration. Add an empty JSON object as the body and the same call succeeds.
A path like data.propertyId resolves to nothing
tawk.to wraps its payload in its own envelope, and the connector passes that whole body through, so paths carry two levels of data. A property list reads data.data[].propertyId and data.data[].name, not data[].propertyId. Every sample output in the palette is modelled that way, so dragging a path from a node's own output preview always produces the correct form. This is the most likely thing to get wrong when writing a path by hand, and the most likely thing for a later editor to simplify and break.
A node shows green but the output contains an error
It should not, and this integration guards against it specifically. tawk.to answers with its own success flag in the body, and its authentication error has been reported arriving without a matching HTTP status, which on a naive client would present as a successful node carrying an error payload. The tawk.to nodes force a failure whenever the body says the call did not succeed, and lift tawk.to's error code and message onto the node output as readable text. Its message names the offending JSON path, such as the exact field a validation rejected, so read that rather than only the status.
Reading a Knowledge Base article returns a server error
This is a known server-side gap on tawk.to rather than a configuration mistake, and the node is configured around it. Asking for the content block has been reported to return a server error while the meta block succeeds, which is why the field selection defaults to meta. Two more of tawk.to's behaviours are reproduced faithfully here rather than papered over: the single-article read requires the category id as well as the article id, and it rejects a real Knowledge Base site id while accepting only the value primary. tawk.to: List Knowledge Base Articles does accept a real site id, so the two nodes legitimately differ.
Creating a Knowledge Base article fails with a bare server error
The usual cause is a slug that already exists. tawk.to requires the article slug to be unique across the Knowledge Base and answers a duplicate with a server error rather than a validation message naming the slug, which is why the slug is a required field on the node rather than something derived from the title: a silently derived duplicate would produce an unexplainable failure. Change the slug and retry. A separate reported fault is worth knowing about if you are sending extra fields: article metadata has been rejected as invalid with tawk.to still investigating, so no metadata field is exposed on the node.
I need to remove a webhook and do not have the hook id
There is no way to look it up through the API, because tawk.to publishes no webhook lister; the only sources are the response from tawk.to: Create Webhook and the tawk.to dashboard. Store the id when you create the subscription, ideally in a Mattering database keyed to the property. Note also what tawk.to: Delete Webhook does when you do run it: there is no restore, a replacement subscription gets a new id, anything keyed to the old id breaks, and deliveries stop immediately, so events occurring after the call are lost rather than queued.
A ticket read is missing the replies, or a chat that is happening right now does not appear
Both are limits of the API rather than of this integration, and tawk.to has confirmed each. A ticket read returns the opening message only; follow-up replies are not exposed, and tawk.to describes adding them as a standing request with no committed timeline. Chat reads are history: in-progress chats cannot be fetched at all, so anything that needs to react during a conversation has to run on a webhook subscription created with tawk.to: Create Webhook. Two neighbours of the same shape, so nobody hunts for them: there is no ticket creation and no contact create or sync on the REST API.

tawk.to is a trademark of its respective owner. Mattering uses platform names to identify the services these integrations connect to; no endorsement or affiliation is implied.

Start Free Explore our Products