> For the complete documentation index, see [llms.txt](https://docs.eesel.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.eesel.ai/apis-and-developer-resources/use-eesel-from-claude-and-other-assistants.md).

# Use eesel from Claude and other AI assistants

Add your eesel workspace to claude.ai, Claude Code, Cursor or any MCP client, and talk to your agent, check on it and change its setup with your dashboard permissions.

Your eesel workspace is an MCP server. Any assistant that speaks MCP can talk to your agent, check on it and change its setup, with the same permissions you have in the dashboard.

## claude.ai (Claude chat, desktop and Cowork)

1. In Claude, open **Settings > Connectors** and choose **Add custom connector**.
2. Name it eesel and enter the server URL: `https://oracle.eesel.app/mcp`
3. Claude opens eesel's sign-in page. Sign in and approve. That's the only step, and it's once.

From then on, in any Claude conversation with the eesel connector on, you can say "ask eesel what it did today", "have eesel draft a reply to the refund ticket", "turn on the Slack automation". Claude calls your agent; the reply is your agent's, with its instructions and knowledge.

What Claude gets is a small set of tools named like the CLI's commands: `chat` (talk to the agent; anything not covered below is something you ask for here), `status`, `integrations`, `automations`, `instructions`, `files`, `activity`, `approvals`, `invite`, `billing`. Editors can change things; viewers get the read side. Anything set to need approval still asks, and shows up in your dashboard as usual. `approvals` lists what is waiting and `decide_approval` approves or denies one, so an editor can clear a hold without leaving Claude.

The connector talks to your workspace's first agent (the oldest one that isn't a supervisor). With several agents, say which one you mean in the message, or use the CLI or Claude Code where the agent is a header.

## Claude Code, Cursor, Codex and other coding agents

For setting eesel up from a repo or a terminal: [Set up from Claude Code or the terminal](/getting-started/set-up-from-a-coding-agent.md). For adding your workspace as an MCP server to Claude Code with a token instead of OAuth: [Claude Code & MCP clients](/apis-and-developer-resources/claude-code-and-mcp-clients.md).

## Any other MCP client

Two ways in, pick whichever your client supports:

* **OAuth**: server URL `https://oracle.eesel.app/mcp`. The client discovers eesel's authorization server from the standard `/.well-known` metadata and opens the sign-in page.
* **A token**: run `npx @eesel/cli mcp token` and use what it prints: the same URL, an `Authorization: Bearer <token>` header, and an `X-Eesel-Agent-Id` header naming the agent. Tokens last 30 days and carry your role at the time they were minted.

## Any AI agent, with no integration at all

An agent that can browse or run commands needs two things to use eesel well:

* <https://www.eesel.ai/llms.txt>, what eesel is and when to use it, in one file.
* <https://www.eesel.ai/agent.txt>, how an agent should set eesel up and call it, including the CLI commands and the rule that authentication always goes through the person's own login.

Every docs page here is also available as markdown by adding `.md` to its URL, and the whole set is at <https://docs.eesel.ai/llms.txt>.

## Related

* [Set up from Claude Code or the terminal](/getting-started/set-up-from-a-coding-agent.md)
* [Use eesel from ChatGPT](/apis-and-developer-resources/use-eesel-from-chatgpt.md)
* [Claude Code & MCP clients](/apis-and-developer-resources/claude-code-and-mcp-clients.md)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.eesel.ai/apis-and-developer-resources/use-eesel-from-claude-and-other-assistants.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
