Dust is an enterprise AI platform that lets teams build custom AI agents connected to internal knowledge and tools. It integrates with 100+ apps like Slack, Google Drive, Notion, GitHub, and Salesforce, giving agents real context about your company. You can create agents without code using natural language instructions, or use the developer API for deeper automation. Dust supports multiple LLMs including GPT-4 and Claude, so you pick the best model for each task. SOC 2 Type II certified with GDPR and HIPAA compliance.

• No-Code Agent Builder: Create custom AI agents in minutes using natural language instructions. Pick data sources, set permissions, and deploy without writing code.
• Extensive Integrations: Connect to Slack, Google Drive, Notion, Confluence, GitHub, Salesforce, Snowflake, BigQuery, Zendesk, and more. Agents pull real-time data from your existing tools.
• Model-Agnostic Architecture: Choose from top LLMs like GPT-4, Claude, Gemini, and Mistral. Switch models per agent based on task requirements.
• Slack-Native Deployment: Deploy agents directly in Slack where your team already works. Mention an agent in any channel to get answers grounded in company data.
• Granular Permissions with Spaces: Control exactly which data each agent can access. Organize agents and data sources into Spaces with role-based access controls.
• Developer API & Custom Actions: Build programmatic workflows with the API. Create custom Dust Apps for advanced automation and integrate via Zapier or Google Sheets.
• Enterprise Security: SOC 2 Type II certified, GDPR and HIPAA compliant. Data is never used for model training. Encryption at rest and in transit by default.
• MCP (Model Context Protocol) Support: Extend agent capabilities with MCP, allowing agents to use external tools and services beyond built-in integrations.

• Fast Agent Creation: Non-technical users can build and customize agents in minutes using the no-code builder, with templates available to speed up common use cases.
• Deep Knowledge Integration: Agents pull real-time data from connected tools, so answers are grounded in your actual company knowledge rather than generic AI responses.
• Strong Security Posture: SOC 2 Type II, GDPR, and HIPAA compliance with fine-grained data access controls give IT teams confidence in deployment.
• Flexible Model Selection: Being model-agnostic means you're not locked into one LLM provider. Pick the best model for each specific use case.
• Active Development: Frequent feature releases and continuous improvements keep the platform evolving quickly.
• Proven Enterprise Adoption: Companies like Doctolib report 70% weekly usage across 3,000 employees, showing the platform scales across large organizations.

• No Permanent Free Tier: After the 14-day trial, the free plan is extremely limited (single user, restricted interactions). At €29/user/month, costs add up quickly for larger teams.
• Learning Curve for Advanced Use: While basic agents are easy to build, designing agents with complex multi-source data, custom actions, and orchestration requires technical onboarding.
• Data Quality Dependency: Agent accuracy depends heavily on the quality of your connected knowledge bases. Outdated or contradictory internal docs lead to unreliable answers.
• Setup for Complex Workflows: Advanced multi-step automation or deeply customized analysis can require additional configuration or external tooling.
• Fair-Use Messaging Limits: Plans advertise unlimited messages, but usage operates under fair-use policies and rate limits that are not fully detailed upfront.

• Customer Support Automation: Connect agents to your help center, product docs, and ticket history. Support teams get instant, accurate answers to customer questions without searching multiple systems. Doctolib achieved 70% weekly usage across 3,000 employees using this approach.
• Sales Enablement & Research: Build agents that pull account snapshots from CRM data, past interactions, and call transcripts. Sales reps can generate targeted outreach, answer RFP questions, and get competitor insights in seconds instead of hours of manual research.
• Internal Knowledge Management: Create a single point of access for company information scattered across Slack, Notion, Google Drive, and Confluence. Employees ask questions in natural language and get answers sourced from verified internal documents.
• Marketing Content Creation: Marketing teams draft brand-consistent content using agents trained on company messaging, product information, and internal documentation.
• Engineering & Ops Workflows: Agents generate incident summaries, automate documentation, and surface relevant knowledge during operational workflows. Companies like Creative Force use Dust to automate administrative and knowledge tasks across teams.
• Legal & Compliance Assistance: Agents help review internal policies, summarize documents, and draft communications while keeping work grounded in approved company knowledge.