Dust is an open-source platform for building and deploying large language model (LLM) agents and custom AI workflows designed for internal team productivity.

• Orchestrates multi-step LLM workflows using a block-based execution engine designed for high reliability and granular observability.
• Synchronizes organizational knowledge via managed connectors for Slack, Notion, Google Drive, and GitHub to facilitate Retrieval-Augmented Generation (RAG).
• Deploys specialized agents through a collaborative interface, enabling team-wide interaction via web applications or direct Slack integration.
• Built with a performance-focused Rust backend and TypeScript frontend to ensure low-latency execution of complex prompt chains.
• Provides model-agnostic support for major providers including OpenAI, Anthropic, and Mistral through standardized API interfaces.

Ideal for automating customer support synthesis, building internal research assistants, and streamlining document-heavy enterprise workflows.