# Dust (slim)

> Dust is a platform for building AI agents on a company's documents and connected tools, from Permutation Labs in Paris. Outside agents reach it through a REST API, a JavaScript SDK, a CLI and a remote MCP server with OAuth.

- Full: https://www.anchorterminal.com/tools/dust.md (~7,050 tokens) · this version ~1,680 tokens · JSON https://www.anchorterminal.com/tools/dust.json · canonical https://www.anchorterminal.com/tools/dust
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-09

**B · 62.8/100 · rank #378 of 842 · #5 in Company knowledge & data catalogues · not agent-ready · confidence medium**


Assessment: The whole platform is MIT on GitHub, with a public OpenAPI document of 68 operations, llms.txt and a remote MCP server that acts with the signed-in user's access. API calls that run agents draw only on a paid workspace credit pool, and the terms of service could not be read.

## Facts

- Kind: HTTP API · vendor: Permutation Labs · category: Company knowledge & data catalogues · legal entity: Permutation Labs · provenance 64/100
- Endpoint: `https://dust.tt/mcp` (HTTP)
- Auth: OAuth or key · pricing: Freemium · x402: no · licence: MIT for the platform's source at github.com/dust-tt/dust. The hosted service is sold under Dust's own terms. The npm page gives ISC for `@dust-tt/client` and MIT for `@dust-tt/dust-cli`
- Probe metrics: not measured yet (probes haven't run)
- REST API: 54 paths and 68 operations under `/api/v1/w/{wId}` on `https://dust.tt` (US) or `https://eu.dust.tt` (EU). Agents, conversations with SSE events, workspace search, data sources, documents, tables, spaces, skills, triggers and analytics export
- MCP server: Remote, at `https://dust.tt/mcp` or `https://eu.dust.tt/mcp`, OAuth only. Lists agents, works with conversations, Pods and files, and searches spaces. It does not proxy the third-party tools set up in the workspace. An admin can switch it off
- Client-side MCP server: Preview. An application registers its own tools for a conversation through the API. OAuth personal access tokens only
- Credentials: Workspace API keys as Bearer tokens, limited to chosen spaces, with an optional spending cap on a rolling 30 days. OAuth with dynamic client registration or client ID metadata documents for MCP
- Rate limits: 120 document upserts a minute per workspace and 10,000 runs a day per Dust app. No figure published for conversations or search
- Search: `GET /api/v1/w/{wId}/search` with `limit` (1 to 100, default 25), `cursor`, `viewType` and `spaceIds`. Data source search takes `top_k`, `full_text`, `target_document_tokens`, tag, parent and timestamp filters
- SDK and CLI: `@dust-tt/client` 1.2.9 (22 September 2026) and `@dust-tt/dust-cli` 0.4.6 (1 September 2026), both on npm and both asking for Node 24.16 or later
- Plans: Free seat 500 lifetime credits. Pro €30 a seat a month or €24 yearly, 8,000 credits. Max €150 or €120 yearly, 40,000 credits. Enterprise by quote, with pooled credits and a programmatic rate
- Credits: Charged per message by model and actions. Tool actions cost 0, 1 or 3 credits by tier, and knowledge search is 3
- Audit: Audit logs on the Enterprise plan only, with user, API key and system actors, CSV export and streaming to Datadog, Splunk, S3, GCS or an HTTPS endpoint
- Certifications: SOC 2 Type II, HIPAA support and GDPR compliance stated on dust.tt/home/security. Vulnerability reports go to a HackerOne programme
- Status: status.dust.tt on Statuspage, 16 components among them API, Conversations, Data Sources, us-central1 and europe-west1

- Scores: Reliability 53, Performance pending, Schema & documentation 75, Agent ergonomics 67, Security & auth 75, Payments & pricing 20, Task success pending, Maintenance & community 86, Transparency & trust 68 · total over the 7 assessed categories
- Why: Reliability, Read with the hosted lines, on the REST API and the remote MCP server. · Schema & documentation, A public OpenAPI 3.0 document with 54 paths and 68 operations, linked from the docs and from llms.txt (25). · Agent ergonomics, Workspace search takes `limit` (1 to 100, default 25), and data source search takes `top_k` and `target_document_tokens`, so results can be… · Security & auth, API keys go in the `Authorization` header, can be limited to chosen spaces and capped by spend, and the MCP server takes OAuth with dynamic… · Payments & pricing, Read with the hosted rubric. · Maintenance & community, The changelog's newest entry is dated 8 October 2026 and the repository had 112 commits that day (30). · Transparency & trust, The platform's source is public under the MIT licence (30).
- Sources: 28, open questions: 10, both in the full twin
- Capabilities: knowledge.search, agent.mcp-client
- JSON: https://www.anchorterminal.com/api/v1/tools/dust.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/dust.svg` or a link to https://www.anchorterminal.com/tools/dust from a page on dust.tt or one of its subdomains, or the README of github.com/dust-tt/dust, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Use `https://eu.dust.tt` for a workspace in the EU region and `https://dust.tt` otherwise, for both the API and the MCP server
2. Connect an MCP client by OAuth and send the `resource` parameter in authorisation and token requests. API keys do not work on the MCP server
3. Expect HTTP 429 with type `rate_limit_error` when the credit pool, the programmatic monthly cap or the key's spend cap is exhausted. Retrying will not help
4. Call `GET /api/v1/w/{wId}/search` with `limit` (1 to 100, default 25) and `cursor`, or a data source view's search with `top_k`
5. Keep document upserts under 120 a minute per workspace

## Connect

```bash
npm install @dust-tt/client
```
Full config and headless snippets are in the full page. Through letme (picks today, calling later): https://letme.dev/dust

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Glean | B | 69.8 | knowledge.search, agent.mcp-client | https://www.anchorterminal.com/tools/glean.min.md |
| Onyx | B | 65 | knowledge.search, agent.mcp-client | https://www.anchorterminal.com/tools/onyx.min.md |
| Cohere North | C | 55.1 | knowledge.search, agent.mcp-client | https://www.anchorterminal.com/tools/cohere-north.min.md |
| Open WebUI | D | 51.8 | agent.mcp-client, knowledge.search | https://www.anchorterminal.com/tools/open-webui.min.md |
| OpenAI Agents SDK | AA | 86.2 | agent.mcp-client | https://www.anchorterminal.com/tools/openai-agents-sdk.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)