# Create scores

URL: https://docs.hanzo.ai/docs/openapi/o11y/createllmscore

> Attaches an eval score or human-feedback signal to a trace or a single observation.

`POST /v1/o11y/llm/scores`

> [O11y →](https://docs.hanzo.ai/docs/openapi/o11y) · [All API references →](https://docs.hanzo.ai/docs/openapi) · [Get started →](https://docs.hanzo.ai/docs/guides)

|               |                                           |
| ------------- | ----------------------------------------- |
| **Address**   | `https://api.hanzo.ai/v1/o11y/llm/scores` |
| **Method**    | `POST`                                    |
| **Operation** | `CreateLLMScore`                          |
| **Auth**      | `Authorization: Bearer $HANZO_API_KEY`    |

Attaches an eval score or human-feedback signal to a trace or a
single observation.

Callers need the editor role; the runtime's own gate enforces it, and it
validates the payload and stamps the score's author and org.

## Request

8 fields, body `application/json` (required).

| Field           | In   | Type            | Required | Description                                                                          |
| --------------- | ---- | --------------- | -------- | ------------------------------------------------------------------------------------ |
| `comment`       | body | string          | —        | Comment is a free-text note.                                                         |
| `dataType`      | body | string          | —        | DataType is the score's kind — NUMERIC or CATEGORICAL.                               |
| `name`          | body | string          | —        | Name is the score's name, e.g. helpfulness.                                          |
| `observationId` | body | string          | —        | ObservationID is the single observation the score attaches to, when narrowed to one. |
| `source`        | body | string          | —        | Source is where the score came from, e.g. API, EVAL.                                 |
| `stringValue`   | body | string          | —        | StringValue is the categorical score, when the score is categorical.                 |
| `traceId`       | body | string          | —        | TraceID is the trace the score attaches to.                                          |
| `value`         | body | number (double) | —        | Value is the numeric score.                                                          |

## Response

| Status    | Body                 | Meaning           |
| --------- | -------------------- | ----------------- |
| `201`     | o11y.O11yLLMScoreOut | created           |
| `202`     | Approval             | held for approval |
| `default` | problem-details      | refused           |

`201` body — 15 fields.

| Field                | In   | Type               | Always | Description                                                            |
| -------------------- | ---- | ------------------ | ------ | ---------------------------------------------------------------------- |
| `data`               | body | o11y.O11yLLMScore  | —      |                                                                        |
| `data.comment`       | body | string             | —      | Comment is a free-text note.                                           |
| `data.createdAt`     | body | string (date-time) | —      | CreatedAt is when the score was stored.                                |
| `data.createdBy`     | body | string             | —      | CreatedBy is who created the score.                                    |
| `data.dataType`      | body | string             | —      | DataType is the score's kind — NUMERIC or CATEGORICAL.                 |
| `data.id`            | body | string             | —      | ID is the score's id.                                                  |
| `data.name`          | body | string             | —      | Name is the score's name.                                              |
| `data.observationId` | body | string             | —      | ObservationID is the observation the score attaches to, when narrowed. |
| `data.source`        | body | string             | —      | Source is where the score came from, e.g. API, EVAL.                   |
| `data.stringValue`   | body | string             | —      | StringValue is the categorical score, when the score is categorical.   |
| `data.timestamp`     | body | string (date-time) | —      | Timestamp is the score's own event time.                               |
| `data.traceId`       | body | string             | —      | TraceID is the trace the score attaches to.                            |
| `data.updatedAt`     | body | string (date-time) | —      | UpdatedAt is when the score last changed.                              |
| `data.value`         | body | number (double)    | —      | Value is the numeric score.                                            |
| `status`             | body | string             | —      | Status is "success".                                                   |

Failure carries the platform error shape — see [Errors](https://docs.hanzo.ai/docs/errors).

## Examples

**CLI**

```bash
hanzo o11y llm scores create
```

**SDK**

**.ts**

```ts
import { Configuration, O11yApi } from 'hanzoai';

const api = new O11yApi(new Configuration({ accessToken: process.env.HANZO_API_KEY }));
const { data } = await api.createLLMScore({ comment: "<comment>", dataType: "<dataType>" });
```

**.py**

```python
from hanzoai.cloud import ApiClient, Configuration
from hanzoai.cloud.api import O11yApi

client = ApiClient(Configuration(access_token=os.environ["HANZO_API_KEY"]))
result = O11yApi(client).create_llm_score(comment="<comment>", data_type="<dataType>")
```

**.go**

```go
cfg := hanzoai.NewConfiguration()
cfg.AddDefaultHeader("Authorization", "Bearer "+os.Getenv("HANZO_API_KEY"))
client := hanzoai.NewAPIClient(cfg)

resp, _, err := client.O11yAPI.CreateLLMScore(context.Background()).Execute()
if err != nil {
	return err
}
```

**.rs**

```rust
use hanzo_client::apis::{configuration::Configuration, o11y_api};

let mut cfg = Configuration::new();
cfg.bearer_access_token = std::env::var("HANZO_API_KEY").ok();

let result = o11y_api::create_llm_score(&cfg, Default::default()).await?;
```

**.java**

```java
import ai.hanzo.cloud.ApiClient;
import ai.hanzo.cloud.api.O11yApi;

ApiClient client = new ApiClient();
client.setBearerToken(System.getenv("HANZO_API_KEY"));

var result = new O11yApi(client).createLLMScore();
```

**HTTP**

```bash
curl -X POST https://api.hanzo.ai/v1/o11y/llm/scores \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "comment": "<comment>",
       "dataType": "<dataType>"
     }'
```

**MCP**

Tool `o11y`, op `CreateLLMScore` — POST the JSON-RPC envelope to `https://api.hanzo.ai/v1/mcp`.

```bash
curl -X POST https://api.hanzo.ai/v1/mcp \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "jsonrpc": "2.0",
       "id": 1,
       "method": "tools/call",
       "params": {
         "name": "o11y",
         "arguments": {
           "op": "CreateLLMScore",
           "input": {
             "comment": "<comment>",
             "dataType": "<dataType>"
           }
         }
       }
     }'
```

---

[O11y API](https://docs.hanzo.ai/docs/openapi/o11y) · [All Hanzo APIs](https://docs.hanzo.ai/docs/openapi) · [Interactive reference](https://docs.hanzo.ai/docs/openapi/reference)