> ## 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.

# Update a scorer

> Use extension__eval__update_scorer through Atlan MCP Gateway.

Call `extension__eval__update_scorer` with the inputs below. Open an MCP session with `mcp__initialise` first and pass its `session_id`. The authenticated `tools/list` response is the final check for the name and schema available to your account.

The update appends a new scorer version. It does not edit an existing version in place.

## Inputs

| Input | Type | Required | Meaning |
| - | - | - | - |
| `backend` | string,null | No | Replace the judge backend; selection is versioned with the definition. Allowed: `classifier`, `llm`, `null`. Default: `null`. |
| `model` | string,null | No | Replace the model catalogue name; a hosted judge cannot clear it. Default: `null`. |
| `name` | string,null | No | Existing slug only; changing it is refused before anything is written. Default: `null`. |
| `display_name` | string,null | No | Default: `null`. |
| `description` | string,null | No | Set or clear the description. Explicit `null` clears it, per RFC 7396. Default: `null`. |
| `messages` | array,null | No | Replace the prompt wholesale. An `llm_judge` cannot clear it. Default: `null`. |
| `output_type` | string,null | No | Change how the judge answers. Changing it usually means changing `choices` in the same patch. Allowed: `classification`, `score`, `null`. Default: `null`. |
| `choices` | array,null | No | Replace the choices wholesale — and with them what the judge may answer. Default: `null`. |
| `state` | object,null | No | Replace or clear what the judge is shown. Explicit `null` clears it, which turns a question-set scorer back into a rubric one only if the same patch supplies a rubric. Default: `null`. |
| `questions` | object,null | No | Replace or clear the questions wholesale. Explicit `null` clears them. Default: `null`. |
| `source_metadata` | object | No | Replace or clear the caller's own metadata record (the envelope's `source_metadata`). The whole object is replaced, not deep-merged; explicit `null` clears it. The gateway reads none of it. Default: `null`. |
| `scorer_id` | string | Yes | The `scorer_…` handle to edit. |
| `if_match` | string | No | The `etag` of your last read; the edit is refused if the scorer has changed since. |
| `rationale` | string | Yes | A concise, first-person explanation of why you are calling this tool right now: what you are trying to accomplish for the user and how this specific call advances that goal. Written for a human auditor reviewing tool usage after the fact. Do not restate the arguments - explain the intent behind them. |
| `session_id` | string | Yes | The id `mcp__initialise` returned for this conversation, so this call is recorded against it. Required: call `mcp__initialise` first and pass the id it returns. Do not invent a value - an id that names no session of yours is refused. |

## Permissions and result

The call uses your authenticated Atlan identity and the target workspace's permissions. This tool changes recorded data. A missing or inaccessible target can be refused even when the tool is listed.

See [MCP tool catalog](/tools/mcp/references/tool-catalog) for related tools and [MCP Gateway errors](/gateway/mcp/troubleshooting) for refusal and connection diagnostics.


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