Mastering OpenClaw | A Quick Guide to Integrating OpenClaw (Clawdbot) into Slack on the Cloud

If you are not yet familiar with OpenClaw (formerly Clawdbot), or if you are familiar with it but have not yet deployed it, we recommend that you first refer to the “Mastering OpenClaw | One-Click, Instant Deployment Guide for OpenClaw (Clawdbot) on the Cloud” to complete the deployment operations before integrating OpenClaw with Slack.

This article mainly introduces how to configure Slack integration after successfully deploying OpenClaw in Tencent Cloud Lighthouse.

Preparatory work

OpenClaw (formerly Clawdbot) natively supports integration with popular chat applications such as WhatsApp, Discord, Telegram, and Slack, allowing interaction with OpenClaw through these applications. Before configuring Slack integration for OpenClaw, please check the following：

You already have one Slack account.
Have you already deployed OpenClaw on a Lighthouse instance (overseas regions recommended)? If not, we recommend:
If you wish to purchase a new Lighthouse server to install and run OpenClaw, you can purchase it from the Tencent Cloud Lighthouse product purchase page or take advantage of a discount during Tencent Cloud’s OpenClaw exclusive promotion. For detailed deployment and configuration procedures for OpenClaw, please refer to the OpenClaw (formerly Clawdbot) One-Click Instant Deployment Guide on the Cloud.
If you created your Lighthouse server using the OpenClaw application template before February 11, 2026, we recommend referring to the updated tutorial: How to Update the Server OpenClaw Application Version.

Create and configure a Slack App

Creating a Slack App

Access the Slack App management page and log in. Click “Create New App” on the page and select “From scratch”.

Enable Socket Mode and generate App Token

In the App configuration page, find Socket Mode in the left navigation bar and click to enable Socket Mode. Only after enabling Socket Mode can the Bot communicate with Slack.

Switch to the Basic Information page in the left navigation bar, scroll down to find App-Level Tokens, and click Generate Token and Scopes.

In the pop-up window, add a Scope (permission) for the token, search for and check connections:write (this is mandatory; otherwise, a connection cannot be established), and click OK.

After generation, copy and save the token (format: xapp-xxxxxxx). Keep it safe and do not disclose it to others, as it will be required for configuration in OpenClaw later.

Generate a Bot Token

In the left navigation bar, find OAuth & Permissions, scroll down to the Scopes module, and click Add an OAuth Scope.

Select Bot scopes and add the following permissions:

After adding permissions, click Install to Workspace at the top of the page. In the pop-up authorization window, confirm the authorization and complete the installation.

After successful installation, the page will automatically generate a Bot User OAuth Token. Immediately copy this token (format: xoxb-xxxxxxx) and save it safely along with the App Token generated in step 1.1, as it will be needed in subsequent steps.

Enable Event Subscriptions

Find Event Subscriptions in the left navigation bar and click Enable Events. Enabling this allows you to receive messages and event notifications from Slack, enabling bot responses.

No further configuration is needed after enabling (subsequent terminal pairings will automatically associate them). Simply keep the page open and continue with the following steps.

Configure the Model for OpenClaw

Next, you need to configure the model for the already deployed OpenClaw instance.

Go to the Tencent Cloud console, select the corresponding Lighthouse instance with deployed OpenClaw, and click the server card to enter the “Manage Instances” page.

Configuring the Model API Key

Configuring the model API Key for OpenClaw can be done directly on the application management page of the Lighthouse server. The panel supports configuring API Keys for commonly used models such as Tencent Hunyuan, Tencent Cloud DeepSeek, Lunar Dark Side Kimi, MiniMax, and Zhipu GLM, as well as other custom models. For details, please refer to the OpenClaw (formerly Clawdbot) One-Click Instant Deployment Guide — Configuring the Model API Key; it will not be elaborated here.

OpenClaw Basic Configuration

First, go to the Lighthouse console to view the OpenClaw instance you just purchased or reinstalled. Click the login button on the page.

After clicking “Login,” on the Tencent Cloud remote terminal (OrcaTerm) page, select “Passwordless Connection” and click “Login.”

The interface after successfully logging into the server via the OrcaTerm remote terminal is shown below:

After successfully logging into the server, enter the following command and press Enter to execute it.

openclaw onboard

After running the openclaw onboard command, you’ll need to complete the subsequent configuration steps using the keyboard. Use the arrow keys to control the options, and press Enter to select and confirm.

Agree to the Disclaimer

After running the above command, a confirmation option will appear: Do you understand the risks? Select “Yes”.

Configuration Mode Selection: Quick Start

Next, you need to select the Onboarding mode. Simply select “QuickStart”.

Configuration Handling

Next, you need to configure Config handling. Select “Use existing values”.

Model Configuration
This step requires selecting the Model/auth provider, i.e., configuring the model. If you have already completed the OpenClaw model configuration in the Tencent Cloud console in the previous steps, you can skip this step. If you have not yet configured the model, you can configure the model API Key via command line in this step.

Kimi：https://platform.moonshot.ai/
MiniMax：https://platform.minimax.io/
GLM：https://z.ai/manage-apikey/apikey-list

As shown in the example below, select Skip for now.

Select All providers.

Select Keep current.

Channel Configuration — Connecting to Slack

Next, proceed to the Select channel step. Use the arrow keys to select Slack (Socket Mode) and press Enter to confirm.

Enter Bot token and App Token

Enter the Bot token and App Token saved in the previous steps, pressing Enter to confirm after each entry.

Configuring Skills

New users do not need to add additional skill packs for now; it is recommended to select “No” here. This is because Skills can enable system-level automation capabilities, including:

File Access
Browser Control
Shell Command Execution
For initial deployments, restricting permissions can improve stability and security.

Enable Hooks (Skip this step if it doesn’t appear).

Recommended to select only: session-memory (Use the up and down arrow keys to move the cursor, the spacebar to select, and the Enter key to confirm and proceed to the next step).

Reason:

No system commands will be executed.
Lowest security risk.
Please do not select:
Execute the script automatically at startup (boot-md)
Command tracing and logging (command-logger)

Installation complete.

Select Restart.

Startup Method

Select “Do this later”.

Install Command Completion Function

It is recommended to select “No”.

Note: Selecting “Yes” here may cause high CPU load on the server during OpenClaw execution, leading to problems with OpenClaw’s operation.

After completing the above steps, wait a moment, and the system will automatically complete the deployment. No additional operation is required during the process. Once the configuration wizard automatically exits and you return to the command-line interface, the deployment is successful. At this point, you have completed the initial integration of OpenClaw with Slack.

Pairing complete

Start Gateway

After successful token verification, log in to the server via a remote terminal, enter the gateway command and press Enter to start the OpenClaw gateway. The gateway will then attempt to establish a connection with the Slack App.

Obtain the pairing code and complete pairing.

After the gateway starts, open the Slack client (or web version). In your configured Slack workspace, send a test message (any content is acceptable) to any channel (or directly to the bot).

After sending the message, Slack will automatically generate a pairing code. Copy this pairing code (note the case sensitivity and ensure no characters are omitted).

Return to the remote terminal and enter the command: openclaw pairing approve slack CODE。

Note: You need to replace code in the command with the pairing code you just received, and do not add the <> symbol when entering it.

After the command is executed successfully, the remote terminal will indicate that the pairing was successful. At this point, OpenClaw and Slack have completed the connection.

Restart the gateway

After successful pairing, enter the gateway command again in the terminal and press Enter to restart the gateway.

Once the gateway starts successfully, the terminal will display “Connection successful.” At this point, you can send messages in Slack, and OpenClaw will respond normally and engage in dialogue using AI. This indicates that the connection is complete.

The result is shown below:

Welcome to join the discussion!

A Discord has been created, and everyone is welcome to join and explore advanced ways to use Openclaw(Clawdbot) together!