Dust
Paris-built enterprise agent platform where people and agents work in shared Pods over the company's own knowledge: 20+ natively maintained connectors plus 70+ MCP, per-agent choice across 20+ models, Spaces-scoped permissions, US or EU residency, and an open-source core.
Dust is a Paris-built enterprise platform for teams that want AI agents grounded in their own company's knowledge instead of general knowledge. It calls its users AI Operators, the people inside a company who build a working AI process themselves, deploy it, and run it for their team.
Agents are built without code. Someone writes what the agent should do in plain language, points it at the data it needs, and picks a model, often starting from a template that Sidekick, Dust's building assistant, fills in. A preview beside the builder shows how the draft answers before it goes live, and developers can add custom tools underneath or manage agents as code from GitHub.
Dust maintains its own connectors to Slack, Notion, Google Drive, Confluence, GitHub and around twenty more, keeping them in sync in real time, and reaches anything else through more than seventy MCP connectors. Agents search across that material semantically, query structured data with SQL, and act back into the same systems.
Work happens in shared space, not in private chats. Pods are persistent workspaces where people and agents collaborate on a project with shared conversations, tasks and files; Frames are interactive dashboards and apps that agents build; and skills are reusable capabilities that improve as teams use them. Agents run on schedules and on events, and multiple agents coordinate on one piece of work.
Every workspace also comes with ready agents, among them @deep-dive for long research across company data and the web, and @analyst for questions about how the team uses Dust. With the memory tool switched on, an agent remembers each person's preferences and context from one conversation to the next, privately, and the person can review or delete what it kept.
Model choice is unrestricted, with more than twenty models from OpenAI, Anthropic, Google, Mistral and DeepSeek selected per agent and none reserved for a higher plan. Usage analytics report by agent and by model, so a team can see what each workflow costs and change the model on that evidence.
Governance is built around Spaces, which segment data and permissions so an agent reaches only what the person asking may see. When a tool would change something in another system, such as publishing in Contentful or editing a sync in Airbyte, the agent stops and asks the user to approve first. On Enterprise, audit logs record each agent run and each tool it used, with the data it touched, and stream to a SIEM such as Datadog or Splunk, alongside SSO, SCIM and custom retention.
Dust is SOC 2 Type II certified and GDPR compliant, encrypts with AES-256 at rest and TLS 1.2+ in transit, does not train models on customer data, and offers US or EU data residency on every plan with single tenant deployment for enterprises. Its core is published on GitHub. Pricing is per seat in euros with a monthly credit allowance, from a free seat to Max at 120€ a month on yearly billing.
From our comparison video
Dust vs Writer (2026): Which Should You Choose?
Our 3 minute video puts Dust side by side with Writer. It covers what each one does and how each is priced, then when to choose which.
Recorded in October 2026. The profile here is live and updates as research is verified.
Vendor details
Canonical URL
https://dust.tt
Category
Enterprise operations agent
Subcategory
Enterprise agent platform — multiplayer workspace over company knowledge
Company status
independent
Use cases & customers
Target customers
Deployment options
Integrations
Two connector tracks. Natively maintained connectors to 20+ data sources, among them Slack, Notion, GitHub, Google Drive and Confluence, are built and kept in sync by Dust rather than routed through a generic integration provider, capped at 3 on the Business plan and unlimited on Enterprise. Alongside them sit 70+ ready made MCP connectors plus native and remote MCP servers for anything else, capped at 5 remote servers on Business. Agents read and act, and credit consumption scales with the tools an agent uses, including search, data retrieval, code execution and actions in connected apps. What a connector exposes is scoped to the Space it belongs to, so an agent reaches only what the asking user may see. Zapier, Make, n8n and Power Automate are listed as tier entitlements, and a Chrome extension brings agents into other tools. Inbound, Dust publishes a Conversation API on Business and adds a Data Source API on Enterprise, with an OpenAPI specification indexed from docs.dust.tt.
In practice
Your team's knowledge is scattered across Notion, Slack, and Drive, and a general chatbot doesn't know any of it. Dust connects its own integrations to those sources so agents answer from your company's actual context.
Someone outside engineering wants to build an agent for their team without waiting on IT. Dust's builder takes plain language instructions, starts from a template with Sidekick's help, and shows the agent's answers in a preview before it goes live, with developer tools available when needed.
Your CISO needs tight control over what AI can touch. Dust separates what an agent can access from who can use it, with permission spaces, SSO, approval before an agent changes data in another system, audit logs that record each tool an agent uses, and a no training commitment.
Sources & related URLs
Research sources
Agentic Index coverage score
12.5 / 14 capabilities · 89%
| Integrations & Tool Calling | Full |
|---|---|
|
Dust offers more than 70 ready made MCP connectors that let agents use other systems' tools. MCP servers, native and remote, are on both plans, with up to 5 remote servers on Business and no limit on Enterprise. Alongside MCP, Dust maintains native connectors to more than 20 data sources, among them Slack, Notion, GitHub, Google Drive, Confluence, Zendesk and Salesforce, with up to 3 on Business and no limit on Enterprise. The native connectors cover the main systems and MCP covers the rest. Agents act as well as read, and credits are charged for tool use that includes actions in connected apps. Zapier, Make, n8n and Power Automate are available on both plans. Each integration belongs to a Space, so what an agent can reach through it follows the customer's own access model, not a blanket service account. Dust's inbound APIs, for outside systems calling Dust, are separate. The 70 and 20 counts are Dust's own, with no published directory listing them. Sourcedust.tt/home/product, /home/pricing feature table and credits FAQread 2026-09-12 |
|
| Workflow Orchestration | Full |
|
Multi agent orchestration with scheduled and event driven triggers is included on both plans, and agents run on those schedules and triggers without a person present. Pods are persistent shared workspaces where people and agents work together on a topic, project or initiative, with shared conversations, tasks, files and agents in one place. Frames are interactive dashboards and apps that agents produce. Pods and Frames run on a shared virtual computer and automations. Work is built from reusable, governable skills, agents and memory that the customer combines into workflows that compound with use. Deep research tasks that need complex, multi step orchestration and tool use consume more credits. A run can involve the people in a Pod, agents working in parallel, the connected systems each step reads and writes, and the model chosen for each agent. Dust does not say what happens when a step in a multi agent run fails or whether it is retried. Sourcedust.tt/home/pricing feature table and /home/productread 2026-09-12 |
|
| Knowledge Grounding & RAG | Full |
|
Agents answer through retrieval augmented generation over a corpus Dust keeps current. Dust maintains its own connectors instead of relying on a generic integration provider, and real time sync keeps agents working from current information. Search, query and extraction across all company data are included on the plans. Semantic search covers unstructured content, and Table Query combines semantic search with SQL for structured data, so one run can answer a numeric question from a warehouse and a policy question from a wiki. Image analysis is also supported. What an agent can retrieve is limited by the asking user's permissions, because data sits in Spaces with configurable access controls and each integration belongs to a Space. Named sources include Slack, Notion, Google Drive, Confluence and GitHub, with more than 20 native connectors in all and anything else over MCP. Dust does not describe chunking, embedding, citations back to a source passage or how fresh a retrieved answer is. Sourcedocs.dust.tt welcome page with dust.tt/home/product and /home/pricingread 2026-09-12 |
|
| Human Oversight & Guardrails | Full |
|
Agents stop and wait for a person before an action that needs approval. Every tool on an MCP server carries a stake level, set in its Tools and Stakes settings, and a tool that asks pauses the agent until the user approves or declines that action, with medium stake approvals remembered per agent. Dust's own connectors come with the gate built in for changes, so publishing or deleting in Contentful, writing to Grafana, changing an Airbyte sync or running a paid Bright Data scraper all wait for a yes. The request reaches people where they are, through a sound alert when an agent is waiting, an email with allow and decline links for agents started by email, and an API call that approves or rejects an action. A credit circuit breaker also pauses an agent at 600 credits on one message and asks whether to continue. Spaces and Groups decide beforehand which data, tools and systems an agent can reach. Sourcedocs.dust.tt changelog (tool stakes and approval entries, January to October 2026) and API reference, validate an actionread 2026-10-21 |
|
| Security, Identity & Governance | Full |
|
Business and Enterprise both carry SOC 2 Type II, GDPR compliance is stated throughout, and HIPAA can be enabled. A Trust Center sits at trust.dust.com, with a vulnerability disclosure page at dust.tt/home/vulnerability. Business includes SSO with Okta, Entra ID and JumpCloud for 5 or more seats, on demand. Enterprise adds SCIM provisioning, audit logs and advanced security controls, plus custom data retention and a custom MSA and DPA, so Business customers have SSO and Spaces but no provisioning or activity trail. On both plans, data is organized into Spaces with configurable access controls, so admins decide which teams can reach which data sources. Integrations with Notion, Slack and Google Drive belong to Spaces, and Groups set permissions. Data is encrypted with AES-256 at rest and TLS 1.2+ in transit. Prompts, documents and outputs are never used to train models, though third party model providers may keep data for up to 30 days for safety and abuse monitoring before deleting it. Sourcedust.tt/home/pricing feature table, /blog/what-is-data-sovereignty, /blog/secure-enterprise-software-ai-powered-teams and docs.dust.tt admin guideread 2026-09-12 |
|
| Observability & Auditability | Full |
|
On Enterprise, audit logs record what agents do as well as what people do. Each agent run, each tool an agent executes and each trigger that starts a run is its own event, and a tool event names the tool, the MCP server and the data sources it reached, and says whether the AI drove the action or a person started it. The log is time ordered and tamper evident, searchable and exportable as CSV, and it streams to Datadog, Splunk, Amazon S3, Google Cloud Storage or any HTTPS endpoint. On every plan, analytics break credit use down by agent, member, model, tool and Pod, and a per call export lists each model call and tool call behind the bill. Sourcedocs.dust.tt Audit Logs, Audit Logs events reference, Analytics and the consumption export APIread 2026-10-21 |
|
| Memory & State Persistence | Full |
|
Agents can keep a memory of each person they work with. Once the Memory tool is added in the builder, the agent decides during a conversation what is worth keeping, and a user can also ask it to remember something, such as a preference for morning meetings. Memories are private to each user and kept separately for each agent, they last until the user deletes them, and the user can browse them in the agent's Memory tab and delete any single item. Within a piece of work, Pods hold the shared conversations, tasks and files for a project, and long conversations are compacted so the agent keeps the thread. Sourcedocs.dust.tt Agent Memory, Tools and Context Compaction pagesread 2026-10-21 |
|
| Deployment & Data Residency | Full |
|
Customers on Business and Enterprise choose US or EU data residency, so a self serve customer at €24 a seat gets the choice. Enterprise adds single tenant deployment. The platform's code is public at github.com/dust-tt/dust, linked from the site footer, and Dust mentions self hosting without documenting how to do it. On EU hosting, data is stored on servers in the European Union, and model inference is routed to EU regions where available, which means not for every model, with Mistral named as the European provider. Dust separates sovereignty, residency and localization, and notes that a European data center run by a US based company can still be compelled to hand data to American authorities. Dust is based in Paris, so that CLOUD Act argument works in its favor, and it says so. Ardabelle, a named customer, chose Dust on EU hosting and GDPR grounds for private equity deal documents. The specific EU region is not named. Sourcedust.tt/blog/what-is-data-sovereignty and /home/pricing feature tableread 2026-09-12 |
|
| Prebuilt Agents, Templates & Packs | Full |
|
Ready agents and a template library come with every workspace. The default agents run as they are and each does its own job: @dust answers from all the company data a user can reach, @deep-dive spends ten to thirty minutes researching across the knowledge base, data warehouses and the web, and @analyst answers questions about how the workspace uses Dust. Admins can switch each one on or off. Templates cover common use cases, and choosing one opens Sidekick with the template loaded, which drafts the instructions and suggests tools and data for the team to adjust before publishing. The workspace gallery then fills with the agents a company's own builders make. Sourcedocs.dust.tt Templates and default agents pages (@dust, @deep-dive, @analyst)read 2026-10-21 |
|
| Triggers & Channel Coverage | Full |
|
Scheduled and event driven triggers come with multi agent orchestration on both Business and Enterprise, and automations help power Pods and Frames. Dust's own connectors sync in real time, so a change in Notion, Slack or GitHub reaches Dust without waiting for a poll, and an agent can be triggered by it. Agents are reached in the Dust workspace, including Pods where people and agents work in parallel, through a Chrome extension that brings agents into the tools people use, and through Slack, Notion, GitHub and Drive. Zapier, Make, n8n and Power Automate are included on both plans and can start a Dust agent from anything they can see. Dust does not list which event types each connector supports or whether customers can combine conditions, and does not explain its automations further. Sourcedust.tt/home/pricing feature table, /home/product and docs.dust.tt welcome pageread 2026-09-12 |
|
| Model Flexibility & Routing | Full |
|
Every plan includes more than 20 frontier models from OpenAI, Anthropic, Google, Mistral and DeepSeek, across the GPT, Claude, Gemini, Mistral and DeepSeek families, open source models among them. The customer chooses the model for each agent. No model is locked behind a higher plan, though more capable models may use more credits per message. Teams can pick the right model for each workflow and switch as their needs change. Mistral AI, a French founded company, is the European option, and for work that must stay under European jurisdiction Dust offers EU hosted infrastructure with model inference in EU regions. Dust presents model choice as a way to avoid vendor lock in and geopolitical risk by controlling where data is processed and which providers are used, and calls being model agnostic a core feature. Usage analytics by agent and by model show what each model costs per workflow. Dust's MCP connectors and its Conversation and Data Source APIs are separate from model routing. Sourcedust.tt/home/pricing model list and FAQ, /home/product, and /blog/what-is-data-sovereigntyread 2026-09-12 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
The platform has its own API. Business includes a Conversation API for driving conversations with agents from outside, and Enterprise adds a Data Source API that writes into the knowledge layer. Platform documentation at docs.dust.tt, linked from the site footer, offers an llms.txt index of every page in Markdown, with endpoints in OpenAPI. The platform's code is public at github.com/dust-tt/dust, and a developer platform at dust.tt/home/solutions/dust-platform covers Dust Apps, described as an extensible, API first platform for custom integration and LLM orchestration. Programmatic use is metered at $0.01 per credit on Business and custom on Enterprise. Zapier, Make, n8n and Power Automate are available on both plans, and a Chrome extension brings agents into other tools. The MCP connectors run the other way, letting Dust call other systems' tools. No endpoint reference or SDK page is cited, and the OpenAPI specification is mentioned but not shown. Sourcedust.tt/home/pricing developer tools table, docs.dust.tt welcome page and site footerread 2026-09-12 |
|
| Testing, Debugging & Optimization | Partial |
|
A builder can try a change before it goes live. A preview area beside the agent builder runs the draft agent on real questions, and the docs advise testing changes there until the answers are consistent. Teams that manage agents as code can keep agent and skill definitions in YAML in a GitHub repository with the official GitHub Action, review edits in pull requests and promote them from a development workspace through staging to production, and changes Sidekick proposes from a conversation arrive as cards to accept or decline. Thumbs up and down on each answer are collected and readable through the API. None of this scores a change, since there are no test cases, pass rates, evaluators or comparisons between two versions of an agent, so judging a change still rests on someone reading the answers. Sourcedocs.dust.tt Troubleshooting your agent, the dust-github-action changelog entry and the feedback APIread 2026-10-21 |
|
| Browser & Computer Use | Not documented |
|
Pods and Frames run on a shared virtual computer and automations, a shared environment where agents run code and produce artifacts, not one where an agent operates a graphical interface. Credits are charged for tools such as search, data retrieval, code execution and actions in connected apps, and interface control is not among them. The Chrome extension lets a person reach Dust agents while browsing, and no agent drives the browser. Every path in and out of Dust is a maintained connector, an MCP server or an API, so a third party redesigning its interface does not affect it, and Dust maintains typed connectors instead of scraping. There is no hosted or local browser session, desktop or remote computer control, RPA, recorder or extension that acts on a page for the agent. Dust does not explain what its shared virtual computer can and cannot do. Sourcedust.tt/home/product, /home/pricing credits FAQ and /home/chrome-extensionread 2026-09-12 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
Dust Default, the platform's default agent, can now create, update and remove any trigger type Dust supports, including scheduled runs and event based automations. Users describe the automation in conversation, such as a summary every Monday morning, and the agent sets up the trigger.
Bears on: Agent capability
View sourceEach Pod in Dust now has its own settings for which network domains its agent sandboxes may reach and which variables and secrets they hold. Workspace admins edit them, and Pod members can view them and request changes.
Bears on: Human approval / guardrails
View sourceDust added controls for credit spend checkpoints on individual agents in Agent Builder and across the workspace in Usage settings. Admins and managers can configure the gate, which pauses an agent after a message consumes at least 600 LLM token credits and asks whether to continue.
Bears on: Human approval / guardrails
View sourcePricing
Free seat at 0€ with 500 lifetime credits; paid entry is Pro at 24€ per seat per month billed yearly (30€ monthly), and Max at 120€ yearly (150€ monthly). All excluding VAT. Enterprise on application. Figures are euros, not dollars.
Per seat per month with a monthly credit allocation attached to each seat, across three seat types inside one self serve Business plan, plus a sales led Enterprise plan. Seats are priced in euros excluding VAT; programmatic API usage is priced in US dollars at $0.01 per credit. Admins assign a seat type per user and can mix and reassign them at any time.
Included quota
Seat types carry monthly credit allocations that reset each billing period: Free 500 credits lifetime, Pro 8,000 credits/month, Max 40,000 credits/month. Seat types can be mixed and reassigned within a workspace. All plans include 20+ models (OpenAI, Anthropic, Google, Mistral, DeepSeek) with per agent model choice and no model gating.
What is public
Dust publishes full pricing, in euros. The Business plan offers three seat types an admin assigns per user: Free (€0, 500 lifetime credits, 3 connectors, 5 Spaces), Pro (€30 a month or €24 a month billed annually, 8,000 credits a month) and Max (€150 a month or €120 a month billed annually, 40,000 credits a month), up to 100 seats, excluding VAT. Every plan includes 20+ models from OpenAI, Anthropic, Google, Mistral and DeepSeek with per agent model choice and no model gating. Enterprise adds unlimited users and connectors, SSO, SCIM, audit logs and advanced credit controls at custom pricing. A 15 day Pro trial and a free option with no card exist.
Billing mechanics
Self serve Business plan with mixable seat types (Free €0, Pro €30 monthly or €24 annual, Max €150 monthly or €120 annual, excluding VAT), each paid seat carrying a monthly credit allocation that resets per billing period; the Free seat's 500 credits are lifetime. Prorated mid cycle seat additions; monthly seats cancel anytime, annual seats stay reassignable until the commitment ends. Enterprise is sales led with unlimited users and connectors, SSO, SCIM, audit logs and advanced credit controls.
Cost watchouts
Credits meter actual usage: higher capability models consume more credits per message, and heavy agent use can exhaust monthly allocations, pushing users to Max seats or admin enabled overage (capped). Business plan caps at 100 seats, 3 connectors, and 5 Spaces; unlimited connectors, SSO, SCIM, and audit logs require the custom priced Enterprise plan.
Variable cost rationale
Credits are charged per message, and the charge scales with the model chosen and the actions the agent takes. Search, data retrieval, code execution and actions in connected apps all draw credits down, and Dust's FAQ says a deep research task with many steps and tool calls uses more. Credits do not roll over, so an unused allowance is lost at the end of each period while a busy month can run out early. The Max seat costs five times the Pro seat and carries five times the credits, which shows that usage, not the seat, is what heavy users pay for. Overage is controlled. Going past a seat's allocation needs an admin to switch on workspace overage, and that overage is capped. Free seats stop and prompt an upgrade, and Enterprise pools credits across the workspace. Programmatic API usage is billed separately at $0.01 per credit, with rate limits as the control for teams that automate against the platform.
Additional watchouts
Costs scale with usage through the credit system rather than just seats, so heavy agent workloads should be modeled on Max seat economics rather than the Pro entry price.
Sales call required
Mixed (some tiers require a call)
Free / trial
Free seat at €0 with 500 lifetime credits and no card, and a 15 day Pro trial
Lowest paid plan
Pro at €24 per seat per month billed yearly, or €30 monthly, with 8,000 credits a month
Commercial notes
Founded by former Stripe engineers, backed by Sequoia, 5,000+ organizations. Positioned as the workspace agent platform for AI Operators; teams like Vanta and Persona cited at 70 to 90 percent team usage.
Key ambiguities
Dust prices in euros. The pricing page reads Pro €24 and Max €120 per seat per month billed yearly, or €30 and €150 monthly, excluding VAT. Seats are billed in euros, while the programmatic API rate is published in dollars at $0.01 per credit. Free is a seat type inside the Business plan, priced at €0 with 500 credits on a lifetime basis rather than monthly, and the Business plan as a whole is "for teams up to 100 people." The three seat types are mixed and matched across one workspace and reassigned as usage changes, an unusual packaging choice and the reason a per seat headline understates the model. Credit mechanics drive the cost profile. A credit is charged per message, scaled by the model used and the actions performed, with search, data retrieval, code execution and actions in connected apps all consuming. Credits do not roll over; each seat's allocation resets every billing period. On running out, a response already generating will finish, then Pro and Max users can continue through workspace overage only if enabled by an admin and only up to a capped amount, while Free users are prompted to upgrade. Enterprise pools credits across the workspace and negotiates volume pricing. The whole self serve plan is priced exactly, in both billing periods, alongside a full feature comparison table, credit mechanics and the programmatic rate; only Enterprise is on application.
Cancellation / refund
Monthly seats cancel anytime, and the subscription runs to the end of the billing period. Annual seats stay reassignable until the commitment ends, and vacated seats are not refunded.
Support SLA / resale
Enterprise adds priority support with an SLA and a dedicated CSM.
Missing data
Enterprise pricing, minimums and the volume pricing structure. The per model credit consumption table: the FAQ explains that a token efficient model consumes few credits and a deep research task consumes many, and points to docs.dust.tt/docs/credits, but publishes no rates, so a buyer cannot forecast credit burn before deploying. The overage cap amount, which is described as capped without a figure. Whether the Free seat's 500 lifetime credits renew under any circumstance. And the USD equivalents, since only the programmatic rate is quoted in dollars.
Related vendors
- 11th Estate — Agentic platform whose AI engine scans markets, matches an…
- Abacum — FP&A platform for growing companies whose AI Modeler builds, edits…
- Adopt AI — Finance agents that categorize, reconcile and close the books inside…
- Aera Technology — Decision intelligence platform where an always on agent executes…
- Akro AI — On premise operational intelligence that automates document heavy…
- alfred_ — AI executive assistant that triages the inbox overnight and scores…
Alternatives to Dust
The closest documented capability profiles to Dust among enterprise operations agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Harvey12.5 / 14Matches Dust across all 14 documented capabilities
- Glean12.0 / 14A lighter documented profile than DustDust vs Glean →
- Salesforce13.0 / 14Fuller documented coverage on Testing, Debugging & Optimization
- Workato12.0 / 14A lighter documented profile than Dust
- Atlassian Rovo13.5 / 14Adds documented Browser & Computer Use
- Boomi12.5 / 14Fuller documented coverage on Testing, Debugging & Optimization
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded