> ## 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 `aireg` CLI; `aireg --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 AI Registry, Atlan's hosted product. The AI registry specification and aireg CLI are maintained separately. /open-source lists publicly reachable signed installers and explains that the source repository and hosted CLI reference currently require GitHub access. Do not claim the repository or its reference docs are publicly accessible. Use these docs for Atlan-specific setup such as the adapter, sign-in and workspaces.

# Agent profile

> Find an agent's definition, sessions, usage, evaluation evidence, relationships, identity, and saved outputs.

Open an agent from its workspace's **Agents** list. Its profile brings together its registered definition and the evidence attributed to it.

| Tab | What it holds | Next action |
| - | - | - |
| **Overview** | The agent's description, instructions, framework, model, and summary of recorded activity | [Edit the agent](/registry/agents/how-tos/edit-agent) |
| **Releases** | Versions and releases, when releases are supported for the agent | Review the saved version and the release state before changing what clients receive |
| **Sessions** | Recorded execution sessions and their events | [Read sessions and events](/registry/agents/concepts/sessions-and-events) |
| **Usage** | Runs, tokens, spend, models, tools, cache use, and recent runs for a selected period | [Inspect a run](/registry/traces/how-tos/inspect-runs) |
| **Evals** | Linked datasets, scorers, and experiment evidence, when Evals is enabled | [Review experiment results](/evals/how-tos/review-experiments) |
| **Relationships** | Recorded connections to skills, files, and other artifacts | Check which context the agent is linked to |
| **Identity** | The agent's credential controls and workspace reach | [Send traces as this agent](/registry/traces/how-tos/send-and-verify) |
| **Library** | Outputs associated with the agent | Open the saved output and its source session when present |

The tabs describe stored records. A linked skill or model does not prove the runtime used it in a particular run; check the trace for that run.

## Identity and access

Each agent authenticates through its own identity. The **Identity** tab shows its client ID when available. **Rotate API key** reveals a new key once; **Rotate client secret** replaces the secret retained by the Gateway. Each action replaces the named credential, so update applications that depend on it.

Access to the agent follows its workspace. **Who can reach it** shows that scope, and **Your permissions** shows the actions available to the signed-in viewer. Credential changes require permission and are enforced by the service.

<Warning>
  **Revoke credentials** invalidates both the API key and client secret on their next request. The app cannot undo this or rotate a revoked identity; new credentials require a new agent. Use rotation when you need to replace a working credential.
</Warning>

## Profile actions

The **More agent actions** menu provides **Edit agent** and **Archive agent**. Editing keeps the machine name and workspace fixed. Archiving removes the agent from lists and stops its key from authenticating; review [Archive an agent](/registry/agents/how-tos/archive-agent) before confirming.


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