Slack is rolling out some exciting new tools that will make AI more integrated into your daily work. These updates include a new real-time search API and a model context protocol (MCP) server. They’re designed to help developers build smarter AI assistants and make it easier for organizations to deploy these agents safely and efficiently.

With these tools, users can get more accurate, relevant information from Slack conversations without leaving the app. Imagine asking a third-party AI like Claude or Perplexity to help prep meeting points or analyze documents right where you’re chatting. This means less app switching and more focus on your work. Rob Seaman, Slack’s chief product officer, highlights how these features will help employees and IT teams alike. Employees get helpful AI right next to their conversations, while IT can deploy agents at scale without risking data security.

Real-Time Search API Brings Conversation Data to Your Fingertips

The new real-time search API lets AI agents access Slack conversations and files instantly. They don’t need to download tons of data or store it somewhere else. Instead, they can search and surface only the relevant messages or files based on what you ask. For example, if you want to find a specific message from a private channel or a direct message, the API handles it seamlessly while respecting access permissions.

This API can connect with tools like Notion or Dropbox Dash. Notion, for instance, can now surface Slack content directly within its workspace, whether it’s in public or private channels, as long as you have access. Dropbox Dash can also show recent Slack conversations, making cross-platform collaboration even easier. The goal is to make conversation data more accessible and useful without compromising security.

The MCP Server Simplifies Developer Work and Enhances AI Accuracy

The MCP server is built on an open-source platform from Anthropic. It standardizes how AI agents access information and perform tasks within Slack. Instead of building complicated, custom integrations for each AI or task, developers can now use a single, consistent protocol that respects user permissions. This cuts down on complexity and speeds up development.

Major AI providers like Anthropic and Perplexity are already using the MCP server. Perplexity’s AI assistant, for example, benefits from grounding its responses in Slack conversations, which makes its answers more relevant and trustworthy. Frank te Pas from Perplexity says this broadens what users can do with AI and boosts confidence in its results.

Currently, both the RTS API and MCP server are in a closed beta, with plans to open to all users by early 2026. However, third-party AI agents are available now through the Slack Marketplace. Pricing will depend on each vendor.

Slack also introduced Work Objects, a tool that helps developers connect other apps to Slack conversation data. It offers “rich app previews,” showing details like file descriptions and images, and lets users take actions—like marking tasks complete—without leaving Slack. This feature will be generally available in late October.

Finally, Slack has launched new developer tools to make building AI agents easier. These include best practices for AI, prebuilt UI components, and updated resources like a CLI for the Bolt framework. Developers can start using these today to create smarter, more integrated AI solutions.

All these innovations are set to make Slack a more powerful hub for collaboration and AI-driven productivity. They aim to bring helpful AI assistants closer to where work happens, streamlining tasks and improving security at the same time.

Inspired by

• https://www.computerworld.com/article/4066301/slack-dev-tools-pave-way-for-new-ai-agents.html

Sources

• gmpg.org
• infoworld.com
• asbpe.org
• eddie-ozzie.com
• podigee.com