> ## Documentation Index
> Fetch the complete documentation index at: https://platform.atlan.com/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> To act on Atlan objects, use the Atlan MCP server at https://api.atlan.com/mcp or the SDK. To install or publish skills from a terminal, use the `brain` CLI; `brain --map json` prints its command map. The `atlanai` CLI is deprecated. Run a read-only identity check before any write.
> The docs MCP server at /mcp searches these docs only. It cannot read or change Atlan objects.
> SDK packages: Python `atlanai` (PyPI) and TypeScript `@atlanai/sdk` (npm). Show Python first, then TypeScript.
> These docs cover Atlan Agent Registry, Atlan's hosted product. The open-source agent registry specification and its CLI (`brain`) are documented, released and maintained in their own repository; /open-source on this site links to it. Answer general specification or `brain` command questions from the project's own docs, and use these docs only for Atlan-specific setup such as the Atlan adapter, sign-in and workspaces.

# Build and equip agents

> Register agents, equip them with governed skills and tools, and connect the clients and frameworks they run in.

Agent Registry gives each agent a governed record, the reviewed skills and tools it can use, and a trail of what it did. It records and equips agents; your runtime, client, or framework runs them.

```text theme={null}
agent record → skills and tools → client or framework → traced runs
```

## Ways to equip agents

| Approach | Use it when | Start at |
| - | - | - |
| **Skills** | An agent needs reviewed, versioned instructions for a job. | [Skills](/registry/skills/index) |
| **Coding clients** | You work in Claude Code or Codex and want governed tools in the session. | [Connect coding clients](/clients/index) |
| **Agent frameworks** | Your agent is built with LangGraph, OpenAI Agents SDK, Google ADK, CrewAI, Pydantic AI, pi, or Vercel AI SDK. | [Supported frameworks](/frameworks/index) |
| **API recipes** | You register agents, publish skills, or publish logical models from code or CI. | [Cookbooks by task](/cookbooks/index) |

## Core offerings

<CardGroup cols={2}>
  <Card title="Bring agent into Atlan" icon="robot" href="/registry/agents/how-tos/bring-agent">
    Register an agent from its repository with its skills and context, and send its first trace.
  </Card>

  <Card title="Find and assess skills" icon="magnifying-glass" href="/registry/skills/how-tos/discover-and-use">
    Find a skill in a workspace and check its version, source, and scan.
  </Card>

  <Card title="Choose skill publishing method" icon="upload" href="/registry/skills/how-tos/publish-and-version">
    Publish skills from Desktop, a terminal, GitHub, or an agent session.
  </Card>

  <Card title="Connect Claude Code" icon="terminal" href="/clients/claude/how-tos/claude-code">
    Give Claude Code governed Atlan tools.
  </Card>
</CardGroup>

## Agent profile

To open an agent, select its workspace, then select **Agent** in **Browse**, or expand **Agents** in the workspace tree. The profile opens on **Overview**. **Sessions** lists recorded execution sessions, and **Usage** shows runs, models, tools, cache use, and recent runs for the selected range. To compare agents across the fleet instead, start from [Compare agent usage](/registry/reporting).

## Choose where to start

| Need | Start at |
| - | - |
| Register an agent without Atlan Desktop | [Create agent in Atlan](/registry/agents/how-tos/create-agent) |
| Use a reviewed skill in Claude now | [Open skill in Claude](/registry/skills/how-tos/use-with-claude) |
| Publish skills from a repository on merge | [Sync skills from GitHub](/registry/skills/how-tos/sync-from-github) |
| Connect Codex | [Connect Codex](/clients/openai/how-tos/codex) |
| Trace a framework agent | [Supported frameworks](/frameworks/index) |
| Register an agent from CI | [Register agents through API](/cookbooks/how-tos/agents) |
| Understand a session's states and events | [How agent sessions work](/registry/agents/concepts/sessions-and-events) |

<Note>
  Every action runs as the signed-in identity and reaches only the workspaces it can access. Review a skill's source and security scan before an agent uses it, and keep credentials in an approved secret store, never in skills, prompts, or configuration files.
</Note>

## See also

* [How workspaces work](/registry/workspaces/index): Scope, membership, and where records live.
* [How trace inspection works](/registry/agents/concepts/trace-inspection): Find execution evidence for an agent or session.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.