Salesforce Inc. today said that collaboration tool Slack is evolving to provide developers with new tools to securely connect artificial intelligence applications to conversational data.
New additions to the developer platform include enhanced developer tools for building, configuring and managing real-time search application programming interfaces, model context protocol servers, and agent AI apps.
As enterprise companies build AI agents, they need to connect them to the daily workflows of employees and customers. Agent AI runs data and access to data is fundamental to its behavior as it provides context and understanding. According to Salesforce, conversation data provides some of the richest contexts for agents to provide personalized assistance.
New real-time search APIs and model context protocol servers allow developers to securely connect to Slack's collaborative spaces and deliver data securely.
The RTS API provides real-time access to conversational data on demand, allowing AI applications to capture current discussions, files, and channels without having to capture everything and sift through it. This reduces the burden of developers in transferring bulk data and storing potentially sensitive data locally and sending it to AI models.
Kurtis Kemple, head of Slack Developer Relations at Salesforce, told SilicOnangle in an interview that the new approach would save time, and that energy developers would spend writing complex code or make multiple API calls to accomplish tasks.
“The workflow between a team of designers, a part of the front engineers and agents in Vercel and Figma – we have a message of 30 or 40 right now,” Kemple says. “I have to make 40 API calls and grab 40 messages. Do I have to do that every time?”
He explained that the new RTS functionality will streamline this process and provide only the conversation data developers needed to feed AI applications, while reducing the number of API calls required. It not only reduces coding complexity, but also reduces overhead and cost.
The MCP server acts as a standard for connecting large language models, AI apps, and agents to contextual information. It provides a unified communication layer between LLM and Slack, reducing the overall coding effort required to connect data sources and tools.
Developed and released as open source by the PBC of Mankind, MCP reduces a single, consistent protocol and fragmented integration that can be tailored to your needs, meeting user specifications. With MCP, developers do not need to manually define all tasks that allow agents to run or implement complex connectors on each LLM.
Additionally, the Slack platform standardizes how developers build and connect dynamic experiences with Slack Work objects. Work objects standardize how third-party data is displayed in Slack, including detailed information, images, documents, and more. Teams can also connect to third parties such as Asana Inc. to take actions and view results without leaving the platform.
The new Agent Developer Tool includes the Vault app's pre-built block kit table and command line interface functionality. Block kit tables are a native way for developers to display structured tabular data in messages that contain columns and rows. By providing them, developers can quickly fill data without defining a message layout. New CLI features enable developers and administrators to efficiently configure and launch applications using scripts and automation tools.
“If you want to build AI, or not, if you build something, it all starts with 'I have an app',” Kemple said. “Developers need to build it.”
Key companies like Anthropic, Google LLC, Perplexity AI Inc., Writer Inc., Dropbox Inc. and Anysphere Inc. Cursors are already building these tools using new AI Slack apps and agents.
For example, Anthropic's Claude Chatbot can be integrated directly into the Slack workspace, allowing teams to communicate directly to the chatbot via messages, AI assistant panels, or @mentions in threads. This includes access to web search, enterprise data sources, and document analytics. Perplexity uses RTS to integrate AI-driven web search capabilities with insights from internal conversations, providing real-time research that intelligently addresses user needs.
“We are thinking about what AI is like and what we need to be able to support in three or four years from now. How will this business line change?” Kemple said. “We want to have a good place for people to come and do those jobs.”
Image: Slack, Salesforce
By engaging with the TheCube community, we support our mission to keep content open and free. Join TheCube's Alumni Trust Networktechnology leaders connect, share intelligence and create opportunities.
- Over 15m viewers of TheCube Videothe power of conversations such as AI, cloud, cybersecurity and more.
- 11.4k+ TheCube Alumni – Connect with over 11,400 technology and business leaders who shape the future through a unique, trustworthy network.
About Silicon Angle Media
Tech Visionarys Founded by John Furrier and Dave Velante, Siliconangle Media has created a dynamic ecosystem of industry-leading digital media brands reaching 15 million elite technical experts. Our new, unique TheCube AI Video Cloud is based on audience interactions. Leveraging TheCubeai.com neural networks, we help technology companies make data-driven decisions and stay at the forefront of industry conversations.
