Inspiration
Slack has evolved beyond a messaging platform into an organization's living knowledge base. Important decisions, technical discussions, project updates, and resolutions are often buried within thousands of messages and threads. While Slack's native search is useful, finding the exact conversation or context behind a decision can still be time-consuming. We built SlackScout to make workplace knowledge instantly discoverable through intelligent, context-aware search.
What it does
SlackScout is an AI-powered Slack knowledge assistant that helps users quickly find relevant conversations, decisions, and information from across their workspace. Using Slack's Real-Time Search (RTS) capabilities and Model Context Protocol (MCP), it performs semantic search instead of relying solely on exact keyword matches.
Users can ask natural language questions, and SlackScout retrieves the most relevant messages, summarizes the findings, and provides direct links to the original conversations for deeper context.
How we built it
We built SlackScout using Node.js and Slack's official SDKs, including:
- @slack/web-api
- @slack/webhook
- @slack/bolt
We created and configured a Slack app, set up the required OAuth scopes and tokens, and connected it to our workspace. The application continuously indexes workspace conversations and leverages RTS and MCP to retrieve relevant content efficiently. We also implemented message formatting, filtering logic, and conversation linking to provide a clean user experience.
Challenges we ran into
One of the biggest challenges was correctly configuring Slack app permissions, OAuth scopes, and authentication flows. Since the application interacts with workspace data, ensuring the right access levels while maintaining security required extensive testing.
We also faced challenges in:
a. Handling different message types and thread structures. b. Filtering bot-generated messages and irrelevant content. c. Formatting search results in a user-friendly way. d. Testing various workspace scenarios to ensure reliable retrieval accuracy.
Accomplishments that we're proud of
We're proud to have built a working end-to-end solution that transforms Slack conversations into a searchable knowledge layer. SlackScout successfully retrieves relevant messages in real time, provides contextual results, and links users directly to the original discussions.
Most importantly, we were able to combine Slack's latest RTS capabilities with MCP to create a practical tool that solves a real workplace productivity problem.
What we learned
Through this project, we gained hands-on experience with Slack's newest platform capabilities, including Real-Time Search and Model Context Protocol integrations. We also learned how to design retrieval workflows, manage workspace permissions, and build scalable search experiences that can operate effectively in production environments.
What's next for SlackScout
Our vision is to evolve SlackScout from a search assistant into a full organizational knowledge agent.
Future enhancements include:
a. AI-generated summaries of retrieved conversations. b. Integration with company-wide AI assistants through MCP. c. Multi-source retrieval from Slack, documentation platforms, and ticketing systems. d. Personalized recommendations based on user context. e. Advanced semantic ranking and decision-tracking capabilities.
Ultimately, we want SlackScout to become the fastest way for employees to discover and understand organizational knowledge, regardless of where it was originally discussed.
Built With
- express.js
- javascript
- mcp
- node.js
- onrender
- slack
- slack-rts
- webhook
Log in or sign up for Devpost to join the conversation.