AI & Models
Anthropic adds interactive apps to Claude
Anthropic launched interactive app integrations for Claude, including Slack and Canva, built on the Model Context Protocol, with Salesforce support expected soon for paid subscribers.
On Monday, Anthropic announced that Claude users can now access interactive apps within the chatbot interface. The new feature is available to Pro, Max, Team, and Enterprise subscribers, but it is not accessible to free users of the platform. Eligible subscribers can activate these tools to enable a logged-in instance of each service directly within Claude. Depending on which of these apps are enabled, users can perform tasks such as sending Slack messages, generating charts, or accessing cloud files. In keeping with Anthropic’s enterprise focus, the initial integrations consist of workplace tools, including:
- Slack
- Canva
- Figma
- Box
- Clay
Additionally, a Salesforce implementation is expected soon. Anthropic stated in a blog post that tasks like data analysis, content design, and project management are more effective when using a dedicated visual interface. The company added that combining these tools with Claude’s intelligence allows users to work and iterate faster than either system could offer on its own. The app integration system is built on the Model Context Protocol (MCP), which is an open standard introduced by Anthropic in 2024 that connects AI assistants to external systems. The setup is similar to OpenAI’s Apps system, which launched in October and enables interactive third-party tools. Both systems of app integrations are built on MCP, which launched support for apps in November.
The new apps are expected to be highly functional when paired with Claude Cowork, which is an agentic tool for multistage tasks that Anthropic launched last week. While apps are not available in Cowork at launch, Anthropic noted that the integration is coming soon. However, the company urged safety and care when deploying these tools, noting that agentic systems can be unpredictable. In its safety documentation for Cowork, the company encourages users to monitor the agent closely and not grant any unnecessary permissions to the system. Specifically, Anthropic warned users to be cautious when granting access to sensitive information, such as financial documents, credentials, or personal records. To mitigate risks, Anthropic recommends that users “Consider creating a dedicated working folder for Claude rather than granting broad access.”
Why it matters
By allowing direct interaction with enterprise tools, Anthropic is evolving Claude from a chatbot into a workflow hub. This integration aims to streamline data analysis and project management tasks directly within the user’s workspace.