Hajra Naz
Anthropic is adding a new way for users to interact with third-party tools inside Claude. Beginning this week, Claude users can launch interactive apps directly within the chatbot interface, the company announced on Monday.
Staying true to Anthropic’s enterprise-first approach, the initial set of apps is focused on workplace tools. At launch, the lineup includes Slack, Canva, Figma, Box, and Clay, with a Salesforce integration expected to follow.
Once enabled, each app connects a logged-in version of the service to Claude, allowing users to perform actions such as sending Slack messages, generating charts, or accessing cloud files — depending on the app.
Read More: Anthropic rolls out Opus 4.5 with Chrome and Excel integrations.
“Analyzing data, designing content, and managing projects all work better with a dedicated visual interface,” Anthropic said in a blog post announcing the feature. “Combined with Claude’s intelligence, you can work and iterate faster than either could offer alone.”
The new apps feature is available to Pro, Max, Team, and Enterprise subscribers, but not to free users. Eligible users can activate apps through the directory at claude.ai/directory.
Anthropic’s approach closely mirrors OpenAI’s Apps system, which launched in October and also enables interactive third-party tools. Both platforms rely on the Model Context Protocol (MCP), an open standard introduced by Anthropic in 2024. Support for app integrations was added to MCP in November, drawing on work from both companies.
The new apps are expected to be especially useful when paired with Claude Cowork, Anthropic’s all-purpose agent tool introduced last week. Built on top of Claude Code, Cowork allows users to assign multistep tasks that draw on large or open-ended datasets — workflows that previously would have required direct terminal access.
Read More: No Code Required: Anthropic Launches Cowork for Claude Code
With app integrations, Cowork could be granted access to cloud storage or active projects, enabling tasks like updating a design in Figma or pulling fresh data from a company’s Box account.
Apps are not available within Cowork at launch, but Anthropic says the integration is “coming soon.”
Anthropic also cautions that agent-based systems can be unpredictable. In its safety documentation for Cowork, the company advises users to closely monitor the agent and limit permissions wherever possible.
“Be cautious about granting access to sensitive information like financial documents, credentials, or personal records,” the company writes. “Consider creating a dedicated working folder for Claude rather than granting broad access.”
