Dust helps enterprises create actionable AI agents with MCP and Claude
Dust,an enterprise AI platform, uses Claude and other frontier models to power customizable AI agents that integrate with company knowledge systems and perform concrete actions across business applications through the Model Context Protocol (MCP).
With Claude, Dust enables:
- AI agents that complete entire workflows automatically, solving the enterprise AI fragmentation challenge
- Enterprise-grade security with Zero Data Retention (ZDR) policies
- Seamless integration across disconnected business applications
- Automated task completion from GitHub issue creation to CRM record updates
Connecting AI across the enterprise
Businesses today struggle with AI tools that operate in silos, unable to share information or coordinate actions. "Companies have various AI tools across departments operating in isolation with no ability to communicate with each other," explains Gabriel Hubert, CEO and co-founder of Dust.
Dust addresses this challenge with a platform that aggregates internal context and leverages Claude's sophisticated reasoning capabilities alongside protocols like MCP. Their solution creates a central AI operating system where agents can access company knowledge and take action across business applications while maintaining enterprise security.
Why Dust chose Claude for enterprise agents
Dust integrates with leading AI providers including Anthropic, selecting Claude specifically for its superior reasoning and multi-step problem-solving capabilities that enterprise agents require. "Claude is a deeply effective partner for a wide range of tasks that involve reasoning and multi-step problem-solving " says Hubert. The platform offers both Claude 3.7 and 4 Sonnet models, with Haiku's faster response times driving significant customer satisfaction.
Security was equally critical in Dust's selection of Claude. Dust implemented enterprise-grade infrastructure with sophisticated permissions designed specifically for MCP-powered agents.
Dust’s platform includes a native permissioning layer (“Groups” & “Spaces”) allowing for a clear separation between data access and agent usage rights. Permission creation, as well as data and tool management is part of the onboarding experience to mitigate sensitive data exposure when AI agents operate across multiple business systems. This is a critical safeguard for the secure deployment of agents powered by internal applications and MCP actions.
Hubert added, "by providing universal AI primitives that make all company workflows more intelligent as well as a proper permissioning system, we are setting the foundations for an agent operating system that is future-proofed and bakes in emerging protocols like MCP.”
Combining Claude with MCP transforms AI from conversation to action
Dust's platform combines leading models like Claude with MCP to create agents that complete tasks independently:
- Creating and assigning issues in GitHub
- Scheduling meetings in calendar applications
- Updating documents and customer records
- Pushing pull requests reviews based on internal coding rules
Hubert explains, "Users want more than just conversational interfaces. Instead of generating a draft, they want to create the actual document automatically. Rather than getting meeting summaries, they need CRM records updated without manual intervention. With our solution, there's no need to switch between multiple applications. Your AI agents can complete entire workflows from start to finish.
Dust's platform now enables teams to automate complete workflows that previously required manual intervention across multiple systems. By eliminating context switching between applications, employees can focus on higher-value activities while AI agents handle routine tasks.
Shaping the future of enterprise AI with Anthropic
Dust sees Claude's reasoning capabilities combined with protocols like MCP as the cornerstone of next-generation enterprise AI. "Agents with the ability to take action and complete tasks are what customers are looking for. MCP is a phenomenal step in that direction and supporting the protocol well is a top priority," states Hubert.
By combining Claude's sophisticated reasoning with MCP's action framework, Dust is building the foundation for what they describe as "superhuman work capabilities"—AI agents that don't just provide information but augment humans to accomplish tasks across enterprise systems.