Salesforce is today announcing the “next evolution” of Slack, giving partners and developers tools to securely connect AI with a rich corpus of customer-owned and -controlled conversational data.
A new real-time search (RTS) API, Model Context Protocol (MCP) server, and enhanced developer tools like prebuilt Block Kit Tables, will now help Slack provide secure, flexible, agent-ready access to conversational data, the cloud giant has revealed. Let’s take a look at what exactly these new features are – and what they mean for the ecosystem.
New Slack Capabilities Explained
Salesforce says that, as companies build agents, there still remains the problem of how best to make them useful in the day-to-day flow of work. Agents run on data, but context is what makes them “powerful”, with the richest context coming from conversations.
“Conversational data is the gold of the agentic era, yet it’s been locked away in unstructured messages and chats, largely out of reach for employees, let alone applications,” Salesforce says in an announcement.
“Enter Slack, the work operating system for the agentic enterprise.”
The context provided by the new tools should allow agents to move beyond generic responses and deliver context-aware, user-specific conversations, making AI more relevant and accurate.
Salesforce mentions big-name companies, including Anthropic, Google, Perplexity, Writer, Dropbox, Notion, Cognition Labs, Vercel, and Cursor, among those who are already building on these capabilities to deliver compelling AI applications and agents, all living natively in Slack.
Securely tapping into Slack’s conversational data means intelligence can be brought directly into the flow of work to accelerate growth, the cloud giant says.
“This isn’t just about integrations; it’s a new way to build and a new way to work,” the company said.
Salesforce says that, by building agents directly in Slack, developers can overcome the problem of endless fragmentation, leading to low app adoption.
There are more than 200,000 SaaS companies in the world, with an average enterprise using over 1,000 applications, resulting in “countless hours lost, with employees clicking between apps, endless copying and pasting, and context switching draining productivity by up to 40%”, the cloud giant says, claiming that it is “no surprise” that nearly half of SaaS licenses go unused as shelfware.
By building AI-powered applications directly in Slack, developers and partners can help organizations unify their tech stack, making Slack “the 2% of your IT budget that maximizes the return on the other 98%”, Salesforce says.
Let’s take a look at what companies can do with this new upgrade:
- Unlock Unstructured Data: Slack is letting organizations securely tap into the vast amounts of collective knowledge stored in Slack conversations to enrich AI and agents.
- Accelerate Decision-Making: By connecting app data, like Agentforce Sales or Workday, with Slack conversations, the platform provides complete context, improving search and AI for faster insights and action.
- Increase Productivity: By integrating apps and agents directly into workflows, Slack eliminates context switching, letting teams cut hours of manual research.
- Solve the ‘Adoption Gap’: Slack brings partner and developer apps directly into the flow of work, helping overcome the challenge of getting employees to actually use them.
- Ensure Security and Compliance: With Slack’s enterprise-grade security, privacy features, and granular permissions, businesses can safeguard data and provide a secure foundation for agentic collaboration, Salesforce says.
Denise Dresser, CEO, Slack, said: “The future of work is undeniably agentic, and the success of AI depends on its seamless integration into human workflows. Our latest Slack platform innovations create the secure, data-rich environment necessary for AI agents to become trusted companions.
“We make it simple for customers and partners to build their AI solutions directly into Slack so that work is more connected, intelligent, and productive than ever before.”
A Deeper Look at the RTS API and MCP Server
Slack’s expanded developer platform is purpose-built for agentic AI, so it should be easy for any company to build powerful, custom AI agents that can access rich, contextual conversational data to get more done, securely.
The real-time search (RTS) API and Model Context Protocol (MCP) server provide secure data access on your terms for developers and third parties:
The RTS API provides apps with real-time access to the rich, contextual conversational data within Slack. By surfacing the most current discussions, files, and channels, the RTS API should make the Slack platform more flexible – and empower AI to act on information in context.
Salesforce says that the RTS API is designed with enterprise security at its core, allowing tools to interact with customers’ conversational data without needing to bulk download it or store it elsewhere.
It respects existing access permissions and retrieves data in a focused, use-case-driven way, returning only the specific messages or files relevant to a query, the cloud giant says.
This should result in AI agents and apps getting the precise, contextual information they need.
The MCP server is meant to simplify and standardize how large language models (LLMs), AI apps, and agents discover contextual information and execute tasks on behalf of Slack users.
With a unified communication layer between the LLM and Slack, it should establish a universal way for AI systems to connect to data sources and tools. This means that fragmented integrations are replaced with a single, consistent protocol that sticks to the user’s specific permissions, and developers don’t have to manually define every possible task an agent can perform or implement complex, service-specific integrations for each LLM.
The RTS API and MCP server transform how partners build on Slack by granting AI-powered applications secure access to real-time conversational data, meaning developers can move beyond basic integrations to create contextual, intelligent experiences that automate complex workflows.
Agents can now search, analyze, and act on Slack data to deliver more accurate, relevant, and personalized outcomes, reducing hallucinations and increasing reliability. These capabilities are not just partner apps in the Slack marketplace. Customers building custom applications for internal use can leverage these capabilities and maintain complete access to their Slack data, with the freedom to use it however they choose.
Additionally, the Slack platform is introducing these new innovations:
- Slack Work Objects: Developers can now build apps that deliver dynamic experiences by connecting app data directly to Slack conversations through rich app previews that include everything from file descriptions to embedded images. This standardizes how third-party data, like detailed information, images, and documents, is displayed in Slack, creating a deeper connection between apps and conversations. Teams can also take action, like marking Asana tasks as complete, without ever having to leave the Slack workspace.
- Agentic Developer Tools: From apps and agents to automated workflows, new capabilities include AI best practices, prebuilt Block Kit Tables, and updated resources like the CLI for Bolt apps so developers have “everything they need” to customize Slack. These upgrades should streamline the entire development lifecycle, from the first line of code to the finished end-user experience.
Third-party solutions are available in the Slack Marketplace, which takes a centralized approach to the control, security, and distribution of apps and agents.
Alongside Slack functionality, the marketplace apps and third-party integrations “play a crucial role in creating a unified ecosystem for human-AI collaboration”, Salesforce says.
The marketplace is meant to make it easier than ever for customers to discover tools that customize and securely extend the functionality of Slack to fit their teams’ needs.
The RTS API and Slack MCP server are currently available in closed beta, with general availability in early 2026.
Third-party AI agents using the capabilities are available on the Slack Marketplace today. Slack Work Objects will be generally available to all developers in late October.
New agentic developer tools are available now.
Final Thoughts
Similar to the announcement earlier this year about Salesforce channels, the cloud giant is clearly trying to provide customers with a single workspace with as few fragmented workflows as possible, meaning users should, in an ideal world, have to work with as few tabs open as possible.
The new developer tools, too, are in this vein of simplifying procedures as much as possible, embedding apps within Slack to make them more appealing for team members to use in everyday workflows.
