No Code AI Slack Bot on Scout
2 minute demo and step-by-step guide to build a Slack AI chatbot
Scout is trusted by hundreds of companies to build secure and compliant conversational AI bots for Slack. Scout’s Slack integration helps teams respond to customer questions, facilitate internal enablement channels, and supercharge productivity. In this guide, we’ll show you a quick video demo, and walk you through the step by step setup instructions to get your AI Slack chatbot up and running on Scout!
🎥 2 Minute AI Slack Bot Demo
👷 Build your own AI Slack Bot on Scout
By following these steps you can rapidly deploy a Slack-based AI chatbot tailored to your unique business or product team needs—whether that’s AI customer support automation, AI chatbots for internal communications and enablement, or AI RFP and security questionnaire response automation. Scout seamlessly brings together everything you need—no separate vector store, no separate orchestrations—so you can focus on delivering an AI-driven solution.
Let’s get started building your no-code AI chatbot with Scout and Slack!
Connect Slack to Scout
- Navigate to the Integrations page and under the Slack option click connect.
- You’ll be prompted to grant permission to Scout; click Allow. If you are not an administrator of the Slack workspace, you may need to work with an admin for this step.
Add Scout to your desired Slack channel
Navigate to the Slack channel you want to integrate the AI Slack bot and @ mention @Scout
to add Scout to that channel.
Build your Knowledge Base in Scout
Whether you’re looking to build an internal knowledge base, enhance customer support automation, or automate rfp and security questionnaire support, providing a robust and relevant set of source material is crucial. By populating Scout with your own content—product manuals, knowledge base articles, previously answered questionnaires or marketing materials—you train your AI Slack bot to give context-relevant answers. This ensures your AI chatbots deliver value right away.
In order for Scout to answer questions the way you want specific to your use case, you’ll need to upload your documents into Scout. Scout stores your documents in what we call collections. There’s a few different ways to upload your relevant data into a Scout collection:
Enter a website or sitemap and scrape a page, or multiple pages. This method is ideal for capturing dynamic content across multiple web pages efficiently.
Upload a file (PDF, CSV, etc.) directly to a Scout collection. This option is perfect for managing static documents and datasets.
Create and edit a blank document directly in the Scout dashboard. This offers a hands-on approach to document creation and customization.
Scout has a robust API, and offers both TypeScript and Python SDKs.
Once you’ve updloaded your relevant data into a Scout collection, you can move on to the next steps.
Deploy the Scout Slack Workflow Template
- Navigate to the Workflows page in the Scout dashboard and click Browse Templates
- Click the Slack category on the left, click the AI Slack Bot (Advanced) template, then click the Use Template button on the bottom right.
Configuring the Slack Workflow Template
- Once the workflow template is deployed, you will be dropped into the workflow editor.
- First, click the Company Name Variable block in the upper right hand corner, and change it from “Scout” to your company’s name.
- Next, you will need to tell Scout which channel it should be listening to in your Slack workspace. You can find your channel ID by selecting the channel name within Slack. Scroll to the bottom channel detail panel to find and copy the channel ID.
- Then, navigate back to the Scout workflow editor, and click the first Slack Message block. Paste your channel ID and press enter.
- Next, you will need to tell the workflow to use the knowledge base collection you setup in step 4. There are two collection blocks in the workflow.
- You will need to update both of these blocks. You need click each of the blocks, click the Collection input, and select your collection for each of them. Once configured, Scout can now perform vector similarity search on your collection to quickly retrieve context-relevant data.
- Lastly, click the Save button in the upper right hand corner to save your changes.
Test your AI Slack Bot
Now you should be good to go! You can test your Scout AI Slack Bot by posting messages in the Slack channel that you’ve configured in the Scout workflow.
Testing validates your entire setup—from the initial Slack integration to the retrieval of content from your knowledge base. It’s also a great time to check for AI chatbot best practices like clarity, tone, and relevance. If something needs tweaking (e.g., retrieving more relevant documents or adjusting your LLM prompts), you can revisit your workflow blocks.
🛟 Need help?
Join our Slack community - our engineering team hangs out there and is happy to help!
➕ Additional Context: How Scout Empowers Your AI Chatbot
What is Scout?
Scout helps you unlock the full potential of AI with automated workflows that bring intelligence to your fingertips. Whether you’re crafting content, enhancing customer support with AI chatbots for customer support, or building the next AI-powered app, Scout makes it simple with no-code AI chatbot builders. With intuitive drag-and-drop functionality, you can easily connect large language models (LLMs), databases, and your tools—no need to manage the tech. Focus on what matters most, and let AI do the rest.
What are Workflows in Scout?
Workflows are the backbone of Scout’s AI automation platform. By configuring blocks—whether it’s an LLM Block to handle text generation or an HTTP Block to fetch data—you can automate tasks like sending Slack notifications, replying to user queries, or conducting advanced RAG for customer support automation. Trigger these workflows directly from Slack, via API, or through Scout’s embeddable Copilot UI.
What are Blocks within a Scout Workflow?
Blocks are modular units within a workflow, each with a specific function. This composable approach lets you build complex logic without writing code. You can chain multiple blocks (e.g., Slack input, LLM processing, document retrieval, and Slack output) for a robust, end-to-end solution.
What is an LLM Block?
The LLM Block integrates OpenAI’s API for chatbots, Anthropic (Claude), and other top-tier LLMs like Llama and Mixtral, enabling AI chatbot solutions with Anthropic models or building AI chatbots with OpenAI. Simply select the model, configure your prompts, and watch Scout’s automation do the rest.
What are Collections in Scout?
Collections form your AI knowledge management system, storing your documents in a vector database. Because Scout automatically sets up and manages the underlying vector database, you only need to focus on the content. This vector similarity search for chatbots ensures your Slack bot can quickly retrieve context-relevant data.