Anthropic launches interactive Claude apps, including Slack and other workplace tools

In a significant update, Anthropic has introduced interactive applications within its chatbot platform, Claude, expanding its utility for users. This new feature, announced on Monday, allows Claude users to access a suite of workplace applications, including popular tools such as Slack, Canva, Figma, Box, and Clay. A Salesforce integration is also on the horizon. Each application enables users to interactively engage with their logged-in services through Claude, facilitating actions like sending messages on Slack, generating visual charts, and accessing cloud storage files, based on the enabled applications. In a blog post detailing the launch, Anthropic emphasized that the combination of dedicated visual interfaces with Claude's advanced intelligence can significantly enhance productivity, enabling quicker work processes and iterations. This feature is available exclusively for Pro, Max, Team, and Enterprise subscribers, with free users currently excluded from access. Eligible users can activate the new tools at claude.ai/directory. The integration of apps resembles the Apps system introduced by OpenAI in October, which also supports interactive third-party tools. Both systems utilize the Model Context Protocol (MCP), an open standard that Anthropic rolled out in 2024, which started supporting apps in November. The new tools are particularly synergistic with Claude Cowork, a multifunctional agent tool released by Anthropic just last week. Cowork, built on Claude Code, enables users to assign complex, multi-stage tasks that leverage extensive datasets—tasks that previously required detailed terminal commands. With the new applications feature, Cowork may soon access cloud files and ongoing projects, allowing for efficient updates, such as modifying a marketing graphic in Figma or utilizing fresh data from a Box account. While apps are not initially available in Cowork, Anthropic has indicated that this integration will be implemented shortly. However, the company advises caution when using agentic systems, as they can exhibit unpredictable behavior. Anthropic's safety documentation for Cowork encourages users to closely monitor the agent and to avoid granting unnecessary permissions, especially concerning sensitive information like financial documents and personal records. As a precaution, they recommend creating a dedicated working folder for Claude to limit access.