> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns.
>
> **Recommended** (the `-y` flags skip prompts, so the command runs without an interactive terminal):
> ```bash
> npx -y @scalekit-inc/cli setup -y
> ```
>
> Global:
> ```bash
> npm install -g @scalekit-inc/cli
> scalekit setup -y
> ```
>
> Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for other Agent Skills-compatible agents in `~/.agents/skills`.
> Skills: integrate-agentkit, implement-saaskit, add-mcp-oauth, implement-sso, implement-scim.
> [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/)

---

# Connect AI agents to DataRobot

The DataRobot connector lets your AI agent act in each user's DataRobot account. Each user connects their own DataRobot access token once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 53 tools.

**Authentication:** Bearer Token
**Categories:** AI, Analytics
**Tools:** 53: 31 read, 15 write, 7 destructive
**Users sign in with:** Access token
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=DATAROBOT)

## Setup

1. ### Install the SDK

   **Node.js**

   ```bash
   npm install @scalekit-sdk/node dotenv
   ```

   **Python**

   ```bash
   pip install scalekit-sdk-python python-dotenv
   ```

2. ### Set your credentials

   Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**.

   ```sh title=".env"
   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>
   ```

3. ### Create the DataRobot connection

   In **AgentKit > Connections**, create a DataRobot connection. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

4. ### Authorize a user and make your first call

   **Node.js** (`quickstart.mts`)

   ```typescript
   import { ScalekitClient } from '@scalekit-sdk/node'
   import 'dotenv/config'
   import { createInterface } from 'node:readline/promises'

   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   )
   const actions = scalekit.actions

   const connector = 'datarobot'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize DataRobot:', link)
   const rl = createInterface({ input: process.stdin, output: process.stdout })
   await rl.question('Press Enter after authorizing...')
   rl.close()

   // Make your first call
   const result = await actions.executeTool({
     connector,
     identifier,
     toolName: 'datarobot_batch_predictions_list',
     toolInput: {},
   })
   console.log(result)
   ```

   ```bash
   npx tsx quickstart.mts
   ```

   **Python** (`quickstart.py`)

   ```python
   import os
   from scalekit import ScalekitClient
   from dotenv import load_dotenv
   load_dotenv()

   scalekit_client = ScalekitClient(
       env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
       client_id=os.getenv("SCALEKIT_CLIENT_ID"),
       client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
   )
   actions = scalekit_client.actions

   connection_name = "datarobot"
   identifier = "user_123"

   # Generate an authorization link for the user
   link_response = actions.get_authorization_link(
       connection_name=connection_name,
       identifier=identifier,
   )
   print("Authorize DataRobot:", link_response.link)
   input("Press Enter after authorizing...")

   # Make your first call
   result = actions.execute_tool(
       tool_input={},
       tool_name="datarobot_batch_predictions_list",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)
   ```

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their DataRobot credentials there. See [Authorize a user](/agentkit/tools/authorize/) for the full flow and statuses.

## Tools

Pass the exact name to `execute_tool`, with an input like each tool's example. To ask for a tool that's missing, use the [request form](https://scalekitsupport.portal.usepylon.com/forms/request-a-connector-tool).

**Node.js**

```typescript
const result = await actions.executeTool({
  toolName: 'datarobot_batch_prediction_get',
  toolInput: {
    prediction_job_id: '<prediction_job_id>',
  },
  connector: 'datarobot',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="datarobot_batch_prediction_get",
    tool_input={
        "prediction_job_id": "<prediction_job_id>",
    },
    connection_name="datarobot",
    identifier="user_123",
)
```

`result.data` is the app's response as JSON, and `result.execution_id` (`executionId` in Node.js) is the ID of the call.

### `datarobot_batch_prediction_get`

Get Batch Prediction Job · Read-only

Get the status, progress, row counts and settings of one batch prediction job.
Returns an object with batchPredictionJobDefinition, created, createdBy, elapsedTimeSec, failedRows, hidden, id, intakeDatasetDisplayName, jobIntakeSize, jobOutputSize, and more.
Use datarobot_batch_prediction_get to poll a job from datarobot_batch_prediction_create.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `prediction_job_id` (`string`, required): Batch prediction job ID. Example: 5e3c94aff86f2d10692497c5

Example input: `{"prediction_job_id":"<prediction_job_id>"}`

### `datarobot_batch_predictions_list`

List Batch Prediction Jobs · Read-only

List batch prediction jobs, filtered by status, deployment or date.
Returns data[] (each with batchPredictionJobDefinition, created, createdBy, elapsedTimeSec, failedRows, hidden, id, intakeDatasetDisplayName) plus count, next, previous, totalCount.
Use datarobot_batch_predictions_list to find jobs. Use datarobot_batch_prediction_get for one job.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cutoff_hours` (`integer`, optional): Only jobs created at most this many hours ago. Example: 24
- `deployment_id` (`string`, optional): Only jobs for this deployment. Example: 5e3c94aff86f2d10692497b6
- `end_date_time` (`string`, optional): Latest creation time, ISO 8601. Example: 2026-11-15T00:00:00Z
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `model_id` (`string`, optional): Only jobs that used this leaderboard model. Example: 5e3c94aff86f2d10692497b8
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sort order which will be applied to batch prediction list. Prefix the attribute name with a dash to sort in descending order, e.g. "-created". One of: created, -created, status, -status. Example: created. One of: `created`, `-created`, `status`, `-status`.
- `start_date_time` (`string`, optional): Earliest creation time, ISO 8601. Example: 2026-11-01T00:00:00Z
- `status` (`string`, optional): Only jobs with this status, such as INITIALIZING, RUNNING, COMPLETED, ABORTED or FAILED. Example: COMPLETED

Example input: `{}`

### `datarobot_dataset_features_list`

List Dataset Features · Read-only

List the features (columns) of a dataset with their types and summary statistics.
Returns data[] (each with dataQualityIssues, dataQualityIssuesTypes, datasetId, datasetVersionId, dateFormat, featureType, id, isZeroInflated) plus count, next, previous, totalCount.
Use datarobot_dataset_features_list for a dataset. Use datarobot_project_features_list for a project.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): Dataset ID, from datarobot_datasets_list. Example: 5e3c94aff86f2d10692497b7
- `featurelist_id` (`string`, optional): Only features in this featurelist. Example: 5e3c94aff86f2d10692497c1
- `include_data_quality` (`string`, optional): If true, include detected data quality issues. Example: true One of: false, true. One of: `false`, `true`.
- `include_plot` (`string`, optional): If true, include histogram data. Example: false One of: false, true. One of: `false`, `true`.
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sort field. One of: featureType, name, id, unique, missing, stddev. Example: name. One of: `featureType`, `name`, `id`, `unique`, `missing`, `stddev`, `mean`, `median`, `min`, `max`, `dataQualityIssues`, `-featureType`, `-name`, `-id`, `-unique`, `-missing`, `-stddev`, `-mean`, `-median`, `-min`, `-max`, `-dataQualityIssues`.
- `search_for` (`string`, optional): Only features whose name contains this text (case-insensitive). Example: age

Example input: `{"dataset_id":"<dataset_id>"}`

### `datarobot_dataset_get`

Get Dataset · Read-only

Get one dataset with its size, row and column counts, processing state and source.
Returns an object with categories, columnCount, createdBy, creationDate, dataEngineQueryId, dataPersisted, dataSourceId, dataSourceType, datasetId, datasetSize, and more.
Use datarobot_dataset_get for one dataset. Use datarobot_dataset_features_list for its columns.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): Dataset ID, from datarobot_datasets_list. Example: 5e3c94aff86f2d10692497b7

Example input: `{"dataset_id":"<dataset_id>"}`

### `datarobot_dataset_projects_list`

List Dataset Projects · Read-only

List the projects built from a dataset.
Returns data[] (each with id, url) plus count, next, previous, totalCount.
Use datarobot_dataset_projects_list to see where a dataset is used.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): Dataset ID, from datarobot_datasets_list. Example: 5e3c94aff86f2d10692497b7
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0

Example input: `{"dataset_id":"<dataset_id>"}`

### `datarobot_dataset_versions_list`

List Dataset Versions · Read-only

List the versions of a dataset.
Returns data[] (each with categories, columnCount, createdBy, creationDate, dataPersisted, datasetId, datasetSize, isDataEngineEligible) plus count, next, previous, totalCount.
Use datarobot_dataset_versions_list to find a datasetVersionId.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): Dataset ID, from datarobot_datasets_list. Example: 5e3c94aff86f2d10692497b7
- `category` (`string`, optional): Only versions with this category. One of: TRAINING, PREDICTION, SAMPLE. Example: TRAINING. One of: `TRAINING`, `PREDICTION`, `SAMPLE`.
- `filter_failed` (`string`, optional): If true, leave out versions whose import failed. Example: true One of: false, true. One of: `false`, `true`.
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sorting order which will be applied to catalog list. One of: created, -created. Example: created. One of: `created`, `-created`.

Example input: `{"dataset_id":"<dataset_id>"}`

### `datarobot_datasets_list`

List Datasets · Read-only

List datasets in the Data Registry that you can access.
Returns data[] (each with categories, columnCount, createdBy, creationDate, dataPersisted, datasetId, datasetSize, isDataEngineEligible) plus count, next, previous, totalCount.
Use datarobot_datasets_list to find a dataset ID. Use datarobot_dataset_get for one dataset.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `category` (`string`, optional): Only datasets with this category. One of: TRAINING, PREDICTION, SAMPLE. Example: TRAINING. One of: `TRAINING`, `PREDICTION`, `SAMPLE`.
- `filter_failed` (`string`, optional): If true, leave out datasets whose import failed. Example: true One of: false, true. One of: `false`, `true`.
- `is_deleted` (`string`, optional): If true, return only soft-deleted datasets. Example: false One of: false, true. One of: `false`, `true`.
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sorting order which will be applied to catalog list. One of: created, -created. Example: created. One of: `created`, `-created`.
- `use_case_ids` (`string`, optional): Only datasets in these Use Cases (IDs, comma-separated). Example: 65a1b2c3d4e5f6a7b8c9d0e1

Example input: `{}`

### `datarobot_deployment_accuracy_get`

Get Deployment Accuracy · Read-only

Get accuracy metrics for a deployment over a period, based on actuals uploaded to DataRobot.
Returns data[] (each with baselineSampleSize, baselineValue, metricName, modelId, percentChange, sampleSize, value) plus batchIds, metrics, modelId, period.
Use datarobot_deployment_accuracy_get to monitor model quality in production.
Accuracy needs actuals and an association ID set on the deployment; check datarobot_deployment_capabilities_get.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `baseline_model_id` (`string`, optional): Model whose metric value is used as the baseline. Example: 5e3c94aff86f2d10692497b8
- `end` (`string`, optional): End of the period, ISO 8601 (default the next hour). Example: 2026-11-15T00:00:00Z
- `metric` (`string`, optional): Name of the metric to retrieve. Must be provided when using multiple modelId query params. One of: AUC, Accuracy, Balanced Accuracy, F1, FPR, FVE Binomial, FVE Gamma, FVE Multinomial, FVE Poisson, FVE Tweedie, Gamma Deviance, Gini Norm, Kolmogorov-Smirnov, LogLoss, MAE, MAPE, MCC, NPV, PPV, Poisson Deviance, R Squared, RMSE, RMSLE, Rate@Top10%, Rate@Top5%, TNR, TPR, Tweedie Deviance, WGS84 MAE, WGS84 RMSE. Example: AUC. One of: `AUC`, `Accuracy`, `Balanced Accuracy`, `F1`, `FPR`, `FVE Binomial`, `FVE Gamma`, `FVE Multinomial`, `FVE Poisson`, `FVE Tweedie`, `Gamma Deviance`, `Gini Norm`, `Kolmogorov-Smirnov`, `LogLoss`, `MAE`, `MAPE`, `MCC`, `NPV`, `PPV`, `Poisson Deviance`, `R Squared`, `RMSE`, `RMSLE`, `Rate@Top10%`, `Rate@Top5%`, `TNR`, `TPR`, `Tweedie Deviance`, `WGS84 MAE`, `WGS84 RMSE`.
- `model_id` (`string`, optional): Model in the deployment (default the champion). Example: 5e3c94aff86f2d10692497b8
- `start` (`string`, optional): Start of the period, ISO 8601 (default 7 days ago). Example: 2026-11-01T00:00:00Z
- `target_class` (`string`, optional): Target class to filter results for. Example: yes

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_deployment_capabilities_get`

Get Deployment Capabilities · Read-only

List which features (drift tracking, accuracy, segmented analysis, challengers and more) a deployment supports.
Returns data[] (each with messages, name, supported).
Use datarobot_deployment_capabilities_get before reading drift or accuracy for a deployment.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_deployment_feature_drift_get`

Get Deployment Feature Drift · Read-only

Get drift scores for the input features of a deployment over a period.
Returns data[] (each with baselineSampleSize, driftScore, featureImpact, name, sampleSize, type) plus batchId, count, metric, modelId.
Use datarobot_deployment_feature_drift_get to find which inputs changed. Use datarobot_deployment_target_drift_get for the prediction.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `end` (`string`, optional): End of the period, ISO 8601 (default the next hour). Example: 2026-11-15T00:00:00Z
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `metric` (`string`, optional): Name of the metric used to calculate the drift. Can be one of psi, kl_divergence, dissimilarity, hellinger, and js_divergence. Defaults to psi. Example: psi. One of: `psi`, `kl_divergence`, `dissimilarity`, `hellinger`, `js_divergence`.
- `model_id` (`string`, optional): Model in the deployment (default the champion). Example: 5e3c94aff86f2d10692497b8
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `start` (`string`, optional): Start of the period, ISO 8601 (default 7 days ago). Example: 2026-11-01T00:00:00Z

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_deployment_features_list`

List Deployment Features · Read-only

List the features a deployment needs for predictions, with type and importance.
Returns data[] (each with dateFormat, featureType, importance, knownInAdvance, name) plus count, next, previous, totalCount.
Use datarobot_deployment_features_list to build a prediction payload for a deployment.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `for_segmented_analysis` (`string`, optional): If true, only features usable for segmented analysis. Example: false One of: false, true. One of: `false`, `true`.
- `include_non_prediction_features` (`string`, optional): If true, return every raw feature in the dataset, not just those needed for prediction. Example: false One of: false, true. One of: `false`, `true`.
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sort order. Example: -importance One of: name, -name, importance, -importance. One of: `name`, `-name`, `importance`, `-importance`.
- `search` (`string`, optional): Only features whose name contains this text. Example: age

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_deployment_get`

Get Deployment · Read-only

Get one deployment with its model, status, importance, prediction server and service, model and accuracy health.
Returns an object with accuracyHealth, approvalStatus, createdAt, creator, defaultPredictionServer, description, governance, hasError, id, importance, and more.
Use datarobot_deployment_get for one deployment. Use datarobot_deployment_service_stats_get for traffic metrics.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_deployment_model_history_get`

Get Deployment Model History · Read-only

List the champion models a deployment has used over time and why each was replaced.
Returns data[] (each with customModelImage, endDate, forecastEndDate, forecastStartDate, modelPackage, predictionWarningBoundaries, replacementReason, startDate) plus count, next, previous, totalCount.
Use datarobot_deployment_model_history_get to audit model replacements.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_deployment_service_stats_get`

Get Deployment Service Health · Read-only

Get service health metrics for a deployment over a period: prediction counts, response and execution times, errors and slow requests.
Returns an object with metrics, modelId, period, segmentAttribute, segmentValue.
Use datarobot_deployment_service_stats_get to check traffic and latency. Use datarobot_deployment_target_drift_get for data drift.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `end` (`string`, optional): End of the period, ISO 8601 (default the next hour). Example: 2026-11-15T00:00:00Z
- `execution_time_quantile` (`number`, optional): Quantile for executionTime metric. Example: 0.5
- `model_id` (`string`, optional): Only metrics for this model in the deployment. Example: 5e3c94aff86f2d10692497b8
- `response_time_quantile` (`number`, optional): Quantile for responseTime metric. Example: 0.5
- `slow_requests_threshold` (`integer`, optional): Threshold for slowRequests metric. Example: 1000
- `start` (`string`, optional): Start of the period, ISO 8601 (default 7 days ago). Example: 2026-11-01T00:00:00Z

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_deployment_target_drift_get`

Get Deployment Target Drift · Read-only

Get the drift score of the predicted target for a deployment over a period.
Returns an object with baselineSampleSize, batchId, driftScore, metric, modelId, period, sampleSize, segmentAttribute, segmentValue, targetName, and more.
Use datarobot_deployment_target_drift_get for prediction drift. Use datarobot_deployment_feature_drift_get for input features.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `end` (`string`, optional): End of the period, ISO 8601 (default the next hour). Example: 2026-11-15T00:00:00Z
- `metric` (`string`, optional): Metric used to calculate drift score. One of: psi, kl_divergence, dissimilarity, hellinger, js_divergence. Example: psi. One of: `psi`, `kl_divergence`, `dissimilarity`, `hellinger`, `js_divergence`.
- `model_id` (`string`, optional): Model in the deployment (default the champion). Example: 5e3c94aff86f2d10692497b8
- `start` (`string`, optional): Start of the period, ISO 8601 (default 7 days ago). Example: 2026-11-01T00:00:00Z

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_deployments_list`

List Deployments · Read-only

List the deployments you can view, filtered by health, status, importance or text.
Returns data[] (each with accuracyHealth, approvalStatus, createdAt, creator, defaultPredictionServer, description, governance, hasError) plus count, next, previous, totalCount.
Use datarobot_deployments_list to find a deployment ID. Use datarobot_deployment_get for one deployment.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `accuracy_health` (`string`, optional): Filters deployments by their accuracy health status. One of: failing, not_started, passing, unavailable, unknown, warning. Example: failing. One of: `failing`, `not_started`, `passing`, `unavailable`, `unknown`, `warning`.
- `created_by_me` (`string`, optional): If true, only deployments you created. Example: true One of: false, true. One of: `false`, `true`.
- `importance` (`string`, optional): Filters deployments by their importance. One of: CRITICAL, HIGH, MODERATE, LOW. Example: CRITICAL. One of: `CRITICAL`, `HIGH`, `MODERATE`, `LOW`.
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `model_health` (`string`, optional): Filters deployments by their model health status. One of: failing, not_started, passing, unavailable, unknown, warning. Example: failing. One of: `failing`, `not_started`, `passing`, `unavailable`, `unknown`, `warning`.
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): The order to sort the deployments.Defaults to order by deployment last prediction timestamp in descending order. One of: label, -label, serviceHealth, -serviceHealth, modelHealth, -modelHealth, accuracyHealth, -accuracyHealth, recentPredictions, -recentPredictions, lastPredictionTimestamp, -lastPredictionTimestamp, currentModelDeployedTimestamp, -currentModelDeployedTimestamp, createdAtTimestamp, -createdAtTimestamp, importance, -importance, fairnessHealth, -fairnessHealth, customMetricsHealth, -customMetricsHealth, actualsTimelinessHealth, -actualsTimelinessHealth, predictionsTimelinessHealth, -predictionsTimelinessHealth. Example: label. One of: `label`, `-label`, `serviceHealth`, `-serviceHealth`, `modelHealth`, `-modelHealth`, `accuracyHealth`, `-accuracyHealth`, `recentPredictions`, `-recentPredictions`, `lastPredictionTimestamp`, `-lastPredictionTimestamp`, `currentModelDeployedTimestamp`, `-currentModelDeployedTimestamp`, `createdAtTimestamp`, `-createdAtTimestamp`, `importance`, `-importance`, `fairnessHealth`, `-fairnessHealth`, `customMetricsHealth`, `-customMetricsHealth`, `actualsTimelinessHealth`, `-actualsTimelinessHealth`, `predictionsTimelinessHealth`, `-predictionsTimelinessHealth`.
- `role` (`string`, optional): Filter deployments to only those that the authenticated user has the specified role for. One of: OWNER, USER. Example: OWNER. One of: `OWNER`, `USER`.
- `search` (`string`, optional): Text to search for in deployment labels and descriptions. Example: churn
- `service_health` (`string`, optional): Filters deployments by their service health status. One of: failing, not_started, passing, unavailable, unknown, warning. Example: failing. One of: `failing`, `not_started`, `passing`, `unavailable`, `unknown`, `warning`.
- `status` (`string`, optional): Filters deployments by their status. One of: active, archived, errored, inactive, launching, replacingModel, stopping. Example: active. One of: `active`, `archived`, `errored`, `inactive`, `launching`, `replacingModel`, `stopping`.

Example input: `{}`

### `datarobot_model_get`

Get Model · Read-only

Get one leaderboard model with its blueprint, featurelist, sample size and metric scores.
Returns an object with blenderModels, blueprintId, dataSelectionMethod, externalPredictionModel, featurelistId, featurelistName, frozenPct, hasCodegen, hasFinetuners, icons, and more.
Use datarobot_model_get for one model. Use datarobot_models_list to compare models.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `model_id` (`string`, required): Model ID, from datarobot_models_list. Example: 5e3c94aff86f2d10692497b8
- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5

Example input: `{"model_id":"<model_id>","project_id":"<project_id>"}`

### `datarobot_models_list`

List Models · Read-only

List the models on a project leaderboard with their scores.
Returns an array of objects with blenderModels, blueprintId, dataSelectionMethod, externalPredictionModel, featurelistId, featurelistName, frozenPct, hasCodegen.
Use datarobot_models_list to compare models. Use datarobot_model_get for one model.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5
- `is_starred` (`string`, optional): If true, only starred models. Example: true One of: false, true. One of: `false`, `true`.
- `name` (`string`, optional): Only models whose type matches this name. Example: eXtreme Gradient Boosted Trees
- `order_by` (`string`, optional): Sort fields, comma-separated; prefix with - for descending. Example: -metric One of: metric, -metric, samplePct, -samplePct. One of: `metric`, `-metric`, `samplePct`, `-samplePct`.
- `sample_pct` (`number`, optional): Only models trained on this sample percentage. Example: 64
- `show_in_sample_scores` (`boolean`, optional): If true, include in-sample scores for models trained into validation. Example: false
- `with_metric` (`string`, optional): Only return scores for this metric. Example: LogLoss

Example input: `{"project_id":"<project_id>"}`

### `datarobot_project_featurelists_list`

List Project Featurelists · Read-only

List the featurelists (named sets of features) in a project.
Returns an array of objects with created, description, features, id, isUserCreated, name, numModels, projectId.
Use datarobot_project_featurelists_list to find a featurelistId for training.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5
- `search_for` (`string`, optional): Only featurelists whose name contains this text. Example: Informative
- `sort_by` (`string`, optional): Sort field. Example: name One of: name, description, features, numModels, created, isUserCreated, -name, -description, -features, -numModels, -created, -isUserCreated. One of: `name`, `description`, `features`, `numModels`, `created`, `isUserCreated`, `-name`, `-description`, `-features`, `-numModels`, `-created`, `-isUserCreated`.

Example input: `{"project_id":"<project_id>"}`

### `datarobot_project_features_list`

List Project Features · Read-only

List the features of a project with type, importance and summary statistics.
Returns an array of objects with dataQualities, dateFormat, featureLineageId, featureType, id, importance, isRestoredAfterReduction, isZeroInflated.
Use datarobot_project_features_list for a project. Use datarobot_dataset_features_list for a Data Registry dataset.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5
- `featurelist_id` (`string`, optional): Only features in this featurelist. Example: 5e3c94aff86f2d10692497c1
- `for_segmented_analysis` (`string`, optional): If true, only features usable for segmented analysis. Example: false One of: false, true. One of: `false`, `true`.
- `search_for` (`string`, optional): Only features whose name contains this text. Example: age
- `sort_by` (`string`, optional): Sort field. Example: importance One of: name, id, importance, featureType, uniqueCount, naCount, mean, stdDev, median, min, max, -name, -id, -importance, -featureType, -uniqueCount, -naCount, -mean, -stdDev, -median, -min, -max. One of: `name`, `id`, `importance`, `featureType`, `uniqueCount`, `naCount`, `mean`, `stdDev`, `median`, `min`, `max`, `-name`, `-id`, `-importance`, `-featureType`, `-uniqueCount`, `-naCount`, `-mean`, `-stdDev`, `-median`, `-min`, `-max`.

Example input: `{"project_id":"<project_id>"}`

### `datarobot_project_get`

Get Project · Read-only

Get one project with its target, metric, partitioning, Autopilot mode and stage.
Returns an object with advancedOptions, autopilotClusterList, autopilotMode, catalogId, catalogVersionId, created, customMetricsLossesInfo, externalTimeSeriesBaselineDatasetMetadata, featureEngineeringPredictionPoint, fileName, and more.
Use datarobot_project_get for full settings. Use datarobot_project_status_get for a quick stage check.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5

Example input: `{"project_id":"<project_id>"}`

### `datarobot_project_job_get`

Get Project Job · Read-only

Get one modeling job that has not finished yet.
Returns an object with id, isBlocked, jobType, message, modelId, projectId, status, url.
Use datarobot_project_job_get to check a job from datarobot_project_jobs_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `job_id` (`string`, required): Job ID. Example: 42
- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5

Example input: `{"job_id":"<job_id>","project_id":"<project_id>"}`

### `datarobot_project_jobs_list`

List Project Jobs · Read-only

List the queued, running and errored modeling jobs of a project.
Returns an object with count, jobs, next, previous.
Use datarobot_project_jobs_list to follow Autopilot. Use datarobot_project_job_get for one job.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5
- `status` (`string`, optional): Only jobs with this status. One of: queue, inprogress, error. Example: inprogress. One of: `queue`, `inprogress`, `error`.

Example input: `{"project_id":"<project_id>"}`

### `datarobot_project_status_get`

Get Project Status · Read-only

Get the stage of a project (for example aim, modeling) and whether Autopilot is done.
Returns an object with autopilotDone, stage, stageDescription.
Use datarobot_project_status_get to poll a project while it is set up or modeled.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5

Example input: `{"project_id":"<project_id>"}`

### `datarobot_projects_list`

List Projects · Read-only

List the modeling projects you can access, optionally filtered by name.
Returns an array of objects with advancedOptions, autopilotClusterList, autopilotMode, created, customMetricsLossesInfo, featureEngineeringPredictionPoint, fileName, holdoutUnlocked.
Use datarobot_projects_list to find a project ID. Use datarobot_project_get for one project.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `feature_discovery` (`string`, optional): If true, return only Feature Discovery projects. Example: false One of: false, true. One of: `false`, `true`.
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sort by name. One of: projectName, -projectName. Example: projectName. One of: `projectName`, `-projectName`.
- `project_id` (`string`, optional): Only these project IDs. Example: 5e3c94aff86f2d10692497b5
- `project_name` (`string`, optional): Only projects with this name. Example: Customer churn

Example input: `{}`

### `datarobot_status_get`

Get Task Status · Read-only

Get the status of an asynchronous task, such as a dataset import or project creation.
Returns an object with code, created, description, message, status, statusId, statusType, type.
Use datarobot_status_get to poll the statusId returned by datarobot_dataset_create_from_url or datarobot_project_create.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `status_id` (`string`, required): Status ID returned by an asynchronous call. Example: 9b7d8e10-3c9e-4b8c-9a0e-2a1b3c4d5e6f

Example input: `{"status_id":"<status_id>"}`

### `datarobot_use_case_datasets_list`

List Use Case Datasets · Read-only

List the datasets linked to a Use Case.
Returns data[] (each with columnCount, createdAt, createdBy, dataPersisted, dataSourceType, dataType, datasetId, datasetSize) plus count, next, previous, totalCount.
Use datarobot_use_case_datasets_list for one Use Case. Use datarobot_datasets_list for the whole Data Registry.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `use_case_id` (`string`, required): Use Case ID, from datarobot_use_cases_list. Example: 65a1b2c3d4e5f6a7b8c9d0e1
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sort order; prefix with - for descending. Example: -createdAt One of: -columnCount, -createdAt, -createdBy, -dataSourceType, -datasetSize, -datasetSourceType, -lastActivity, -modifiedAt, -modifiedBy, -name, -rowCount, columnCount, createdAt, createdBy, dataSourceType, datasetSize, datasetSourceType, lastActivity, modifiedAt, modifiedBy, name, rowCount. One of: `-columnCount`, `-createdAt`, `-createdBy`, `-dataSourceType`, `-datasetSize`, `-datasetSourceType`, `-lastActivity`, `-modifiedAt`, `-modifiedBy`, `-name`, `-rowCount`, `columnCount`, `createdAt`, `createdBy`, `dataSourceType`, `datasetSize`, `datasetSourceType`, `lastActivity`, `modifiedAt`, `modifiedBy`, `name`, `rowCount`.
- `search` (`string`, optional): Only datasets whose name matches this text. Example: customers

Example input: `{"use_case_id":"<use_case_id>"}`

### `datarobot_use_case_deployments_list`

List Use Case Deployments · Read-only

List the deployments linked to a Use Case.
Returns data[] (each with createdAt, createdBy, description, id, label, lastActivity, type, updatedAt) plus count, next, previous, totalCount.
Use datarobot_use_case_deployments_list for one Use Case. Use datarobot_deployments_list for every deployment.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `use_case_id` (`string`, required): Use Case ID, from datarobot_use_cases_list. Example: 65a1b2c3d4e5f6a7b8c9d0e1
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sort order; prefix with - for descending. Example: -createdAt One of: -createdAt, -createdBy, -lastActivity, -name, -updatedAt, -updatedBy, createdAt, createdBy, lastActivity, name, updatedAt, updatedBy. One of: `-createdAt`, `-createdBy`, `-lastActivity`, `-name`, `-updatedAt`, `-updatedBy`, `createdAt`, `createdBy`, `lastActivity`, `name`, `updatedAt`, `updatedBy`.
- `search` (`string`, optional): Only deployments whose name matches this text. Example: churn

Example input: `{"use_case_id":"<use_case_id>"}`

### `datarobot_use_case_get`

Get Use Case · Read-only

Get one Use Case with its description and counts of projects, datasets, applications and deployments.
Returns an object with advancedTour, applicationsCount, created, createdAt, createdBy, customApplicationSourcesCount, customApplicationsCount, customJobsCount, customModelVersionsCount, datasetsCount, and more.
Use datarobot_use_case_get for one Use Case. Use datarobot_use_cases_list to browse.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `use_case_id` (`string`, required): Use Case ID, from datarobot_use_cases_list. Example: 65a1b2c3d4e5f6a7b8c9d0e1

Example input: `{"use_case_id":"<use_case_id>"}`

### `datarobot_use_case_projects_list`

List Use Case Projects · Read-only

List the modeling projects in a Use Case.
Returns data[] (each with autopilotDataSelectionMethod, created, createdAt, createdBy, dataset, datasetId, datasetVersionId, hasHoldout) plus count, next, previous, totalCount.
Use datarobot_use_case_projects_list for one Use Case. Use datarobot_projects_list for every project.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `use_case_id` (`string`, required): Use Case ID, from datarobot_use_cases_list. Example: 65a1b2c3d4e5f6a7b8c9d0e1
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sort order; prefix with - for descending. Example: -createdAt One of: -createdAt, -createdBy, -dataset, -featureCount, -fullName, -lastActivity, -models, -name, -projectId, -rowCount, -target, -targetType, -timeAware, -updatedAt, -updatedBy, createdAt, createdBy, dataset, featureCount, fullName, lastActivity, models, name, projectId, rowCount, target, targetType, timeAware, updatedAt, updatedBy. One of: `-createdAt`, `-createdBy`, `-dataset`, `-featureCount`, `-fullName`, `-lastActivity`, `-models`, `-name`, `-projectId`, `-rowCount`, `-target`, `-targetType`, `-timeAware`, `-updatedAt`, `-updatedBy`, `createdAt`, `createdBy`, `dataset`, `featureCount`, `fullName`, `lastActivity`, `models`, `name`, `projectId`, `rowCount`, `target`, `targetType`, `timeAware`, `updatedAt`, `updatedBy`.
- `search` (`string`, optional): Only projects whose name matches this text. Example: churn

Example input: `{"use_case_id":"<use_case_id>"}`

### `datarobot_use_cases_list`

List Use Cases · Read-only

List the Use Cases (workspaces that group projects, datasets and deployments) you can access.
Returns data[] (each with advancedTour, applicationsCount, created, createdAt, createdBy, customApplicationSourcesCount, customApplicationsCount, customJobsCount) plus count, next, previous, totalCount.
Use datarobot_use_cases_list to find a Use Case ID. Use datarobot_use_case_get for one Use Case.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `created_by` (`string`, optional): Filter Use Cases to return only those created by the selected user. Example: 5e3c94aff86f2d10692497a1
- `entity_id` (`string`, optional): The id of the entity type that is linked with the Use Case. Example: 5e3c94aff86f2d10692497b7
- `entity_type` (`string`, optional): The entity type that is linked to the Use Case. One of: project, dataset, file, notebook, application, recipe, playground, vectorDatabase, syftrSearchInstance, embeddingModelValidation, customModelVersion, registeredModelVersion, deployment, customApplication, customApplicationSource, customJob, workload, artifactRepository, workloadStorage, promptTemplate, customModel, registeredModel, enclave, executionEnvironment, entityNotificationPolicyTemplate, notificationChannelTemplate. Example: project. One of: `project`, `dataset`, `file`, `notebook`, `application`, `recipe`, `playground`, `vectorDatabase`, `syftrSearchInstance`, `embeddingModelValidation`, `customModelVersion`, `registeredModelVersion`, `deployment`, `customApplication`, `customApplicationSource`, `customJob`, `workload`, `artifactRepository`, `workloadStorage`, `promptTemplate`, `customModel`, `registeredModel`, `enclave`, `executionEnvironment`, `entityNotificationPolicyTemplate`, `notificationChannelTemplate`.
- `limit` (`integer`, optional): Maximum number of results to return. Example: 20
- `offset` (`integer`, optional): Number of results to skip. Example: 0
- `order_by` (`string`, optional): Sort order; prefix a field with - for descending. Example: -createdAt One of: -applicationsCount, -createdAt, -createdBy, -customApplicationsCount, -datasetsCount, -description, -filesCount, -id, -name, -notebooksCount, -playgroundsCount, -potentialValue, -projectsCount, -stage, -updatedAt, -updatedBy, -vectorDatabasesCount, applicationsCount, createdAt, createdBy, customApplicationsCount, datasetsCount, description, filesCount, id, name, notebooksCount, playgroundsCount, potentialValue, projectsCount, stage, updatedAt, updatedBy, vectorDatabasesCount. One of: `-applicationsCount`, `-createdAt`, `-createdBy`, `-customApplicationsCount`, `-datasetsCount`, `-description`, `-filesCount`, `-id`, `-name`, `-notebooksCount`, `-playgroundsCount`, `-potentialValue`, `-projectsCount`, `-stage`, `-updatedAt`, `-updatedBy`, `-vectorDatabasesCount`, `applicationsCount`, `createdAt`, `createdBy`, `customApplicationsCount`, `datasetsCount`, `description`, `filesCount`, `id`, `name`, `notebooksCount`, `playgroundsCount`, `potentialValue`, `projectsCount`, `stage`, `updatedAt`, `updatedBy`, `vectorDatabasesCount`.
- `project_id` (`string`, optional): Only return experiment containers associated with the given project ID. Example: 5e3c94aff86f2d10692497b5
- `search` (`string`, optional): Only Use Cases whose name matches this text. Example: churn
- `stage` (`string`, optional): Only return Use Cases in the given stage. Example: experimentation

Example input: `{}`

### `datarobot_batch_prediction_create`

Create Batch Prediction Job · Write

Score a dataset with a deployment as a batch prediction job, reading from and writing to the sources you set.
Returns an object with batchPredictionJobDefinition, created, createdBy, elapsedTimeSec, failedRows, hidden, id, intakeDatasetDisplayName, jobIntakeSize, jobOutputSize, and more.
Use datarobot_batch_prediction_create for bulk scoring. Follow it with datarobot_batch_prediction_get.
Use a Data Registry dataset or a cloud/database source as intake; local file upload is not available through this tool.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `deployment_id` (`string`, required): Deployment to score with. Example: 5e3c94aff86f2d10692497b6
- `intake_settings` (`object`, required): Where to read the data. For a Data Registry dataset use {"type": "dataset", "datasetId": "<id>"}; cloud storage (s3, gcp, azure) and database (jdbc, snowflake, bigquery) sources take their own fields. Example: {"type": "dataset", "datasetId": "5e3c94aff86f2d10692497b7"}
- `output_settings` (`object`, required): Where to write the predictions, for example {"type": "s3", "url": "s3://bucket/predictions.csv", "credentialId": "<id>"}. Example: {"type": "s3", "url": "s3://my-bucket/predictions.csv", "credentialId": "5e3c94aff86f2d10692497c4"}
- `abort_on_error` (`boolean`, optional): Should this job abort if too many errors are encountered. Example: true
- `include_prediction_status` (`boolean`, optional): Include prediction status column in the output. Example: false
- `include_probabilities` (`boolean`, optional): Include probabilities for all classes. Example: true
- `max_explanations` (`integer`, optional): Number of prediction explanations per row (0 for none). Example: 3
- `passthrough_columns` (`array`, optional): Columns from the input to copy into the output. Example: ["customer_id"]
- `passthrough_columns_set` (`string`, optional): Set to all to copy every input column into the output. Example: all One of: all. One of: `all`.
- `prediction_threshold` (`number`, optional): Threshold is the point that sets the class boundary for a predicted value. the model classifies an observation below the threshold as false, and an observation above the threshold as true. in other words, datarobot automatically assigns the positive class label to any prediction exceeding the threshold. this value can be set between 0.0 and 1.0. Example: 0.5
- `skip_drift_tracking` (`boolean`, optional): Skip drift tracking for this job. Example: false

Example input: `{"deployment_id":"<deployment_id>","intake_settings":{"type":"dataset","datasetId":"5e3c94aff86f2d10692497b7"},"output_settings":{"type":"s3","url":"s3://my-bucket/predictions.csv","credentialId":"5e3c94aff86f2d10692497c4"}}`

### `datarobot_dataset_create_from_url`

Create Dataset from URL · Write

Import a dataset into the Data Registry from a public or presigned URL. The import runs asynchronously.
Returns an object with catalogId, catalogVersionId, statusId.
Use datarobot_dataset_create_from_url to load CSV, Parquet or similar files that are reachable by URL.
Poll datarobot_status_get with the returned statusId until the import completes.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `url` (`string`, required): URL of the file to import. Example: https://s3.amazonaws.com/datarobot_public_datasets/10k_diabetes.csv
- `categories` (`string`, optional): Intended use of the dataset. One of: TRAINING, PREDICTION. Example: TRAINING. One of: `TRAINING`, `PREDICTION`.
- `do_snapshot` (`boolean`, optional): If true (default), DataRobot stores a snapshot of the data; if false it reads the source each time. Example: true
- `persist_data_after_ingestion` (`boolean`, optional): If true, keep the data after it is ingested. Example: true

Example input: `{"url":"<url>"}`

### `datarobot_dataset_update`

Update Dataset · Write

Rename a dataset or change its categories.
Returns an object with categories, createdBy, creationDate, dataPersisted, datasetId, isDataEngineEligible, isLatestVersion, isSnapshot, name, processingState, and more.
Use datarobot_dataset_update to edit dataset metadata.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): Dataset ID, from datarobot_datasets_list. Example: 5e3c94aff86f2d10692497b7
- `categories` (`array`, optional): Dataset categories. Example: ["TRAINING"] One of: BATCH_PREDICTIONS, MULTI_SERIES_CALENDAR, PREDICTION, SAMPLE, SINGLE_SERIES_CALENDAR, TRAINING.
- `name` (`string`, optional): New dataset name. Example: Customers 2026

Example input: `{"dataset_id":"<dataset_id>"}`

### `datarobot_deployment_create`

Create Deployment · Write

Deploy a registered model package.
Returns an object with id.
Use datarobot_deployment_create after datarobot_model_package_create.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `description` (`string`, required): Deployment description. Example: Scores customers nightly
- `label` (`string`, required): Deployment name. Example: Churn model (production)
- `model_package_id` (`string`, required): Model package ID from datarobot_model_package_create. Example: 5e3c94aff86f2d10692497c2
- `default_prediction_server_id` (`string`, optional): ID of the prediction server to use; required on managed cloud accounts that use dedicated prediction servers. Example: 5e3c94aff86f2d10692497c3
- `importance` (`string`, optional): Business importance of the deployment. Example: HIGH One of: CRITICAL, HIGH, MODERATE, LOW. One of: `CRITICAL`, `HIGH`, `MODERATE`, `LOW`.
- `status` (`string`, optional): Status of the deployment. One of: active, inactive. Example: active. One of: `active`, `inactive`.

Example input: `{"description":"<description>","label":"<label>","model_package_id":"<model_package_id>"}`

### `datarobot_deployment_status_update`

Activate or Deactivate Deployment · Write

Activate or deactivate a deployment. A deactivated deployment stops serving predictions.
Returns the JSON response from the API.
Use datarobot_deployment_status_update to pause or resume a deployment without deleting it.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `status` (`string`, required): New status. Example: inactive One of: active, inactive. One of: `active`, `inactive`.

Example input: `{"deployment_id":"<deployment_id>","status":"active"}`

### `datarobot_deployment_update`

Update Deployment · Write

Change the label, description or importance of a deployment.
Returns the JSON response from the API.
Use datarobot_deployment_update to edit deployment details. Use datarobot_deployment_status_update to activate or deactivate it.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `description` (`string`, optional): New description. Example: Scores customers hourly
- `importance` (`string`, optional): Shows how important this deployment is. One of: CRITICAL, HIGH, MODERATE, LOW. Example: CRITICAL. One of: `CRITICAL`, `HIGH`, `MODERATE`, `LOW`.
- `label` (`string`, optional): New deployment name. Example: Churn model v2

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_model_package_create`

Register Leaderboard Model · Write

Register a leaderboard model as a model package in the Model Registry, ready to deploy.
Returns the new model package, including its id.
Use datarobot_model_package_create before datarobot_deployment_create.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `model_id` (`string`, required): Leaderboard model ID. Example: 5e3c94aff86f2d10692497b8
- `description` (`string`, optional): Model package description. Example: Best model from the Q4 churn project
- `name` (`string`, optional): Model package name. Example: Churn XGBoost v1
- `prediction_threshold` (`number`, optional): Class boundary for binary classification, 0 to 1. Example: 0.5

Example input: `{"model_id":"<model_id>"}`

### `datarobot_model_train`

Train Model · Write

Train a new model on a project from a blueprint, featurelist and sample size. Training runs asynchronously.
Returns the JSON response from the API.
Use datarobot_model_train to add a specific model to the leaderboard after Autopilot.
Follow the job with datarobot_project_jobs_list.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `blueprint_id` (`string`, required): Blueprint ID, from the blueprintId of a leaderboard model. Example: d4c06a5c23cf1d917019720bceba32c8
- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5
- `featurelist_id` (`string`, optional): Featurelist to train on (default the project featurelist). Example: 5e3c94aff86f2d10692497c1
- `sample_pct` (`number`, optional): Percentage of rows to train on. Give this or trainingRowCount. Example: 64
- `scoring_type` (`string`, optional): Validation type for scoring. Example: validation One of: validation, crossValidation. One of: `validation`, `crossValidation`.
- `training_row_count` (`integer`, optional): Number of rows to train on. Give this or samplePct. Example: 5000

Example input: `{"blueprint_id":"<blueprint_id>","project_id":"<project_id>"}`

### `datarobot_project_autopilot_control`

Pause or Resume Autopilot · Write

Pause or resume Autopilot for a project.
Returns the JSON response from the API.
Use datarobot_project_autopilot_control to control a running Autopilot.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `command` (`string`, required): stop pauses Autopilot so no new jobs start; start resumes it and runs queued jobs. One of: start, stop. Example: stop. One of: `start`, `stop`.
- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5

Example input: `{"command":"start","project_id":"<project_id>"}`

### `datarobot_project_clone`

Clone Project · Write

Create a copy of a project; the copy starts data analysis again and is ready for a new target.
Returns an object with pid.
Use datarobot_project_clone to try different settings on the same data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `project_id` (`string`, required): ID of the project to clone. Example: 5e3c94aff86f2d10692497b5
- `copy_options` (`boolean`, optional): If true, also copy the advanced options of the original project. Example: false
- `project_name` (`string`, optional): Name for the copy. Example: Customer churn (copy)

Example input: `{"project_id":"<project_id>"}`

### `datarobot_project_create`

Create Project · Write

Create a modeling project from a Data Registry dataset or a file URL. DataRobot then runs exploratory data analysis asynchronously.
Returns an object with pid.
Use datarobot_project_create to start modeling, then datarobot_project_target_set to start Autopilot.
Poll datarobot_status_get with the returned statusId (or datarobot_project_status_get) until the project is ready for a target.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dataset_id` (`string`, optional): ID of a Data Registry dataset to build the project from. Give this or url. Example: 5e3c94aff86f2d10692497b7
- `dataset_version_id` (`string`, optional): Specific dataset version to use (default the latest). Example: 5e3c94aff86f2d10692497b9
- `project_name` (`string`, optional): Project name. Example: Customer churn
- `recipe_id` (`string`, optional): ID of a wrangling recipe to build the project from. Example: 65a1b2c3d4e5f6a7b8c9d0f2
- `url` (`string`, optional): URL of a data file to build the project from. Give this or datasetId. Example: https://s3.amazonaws.com/datarobot_public_datasets/10k_diabetes.csv

Example input: `{}`

### `datarobot_project_target_set`

Set Target and Start Modeling · Write

Set the target for a project and start modeling (Autopilot) in the chosen mode.
Returns the JSON response from the API.
Use datarobot_project_target_set once the project has finished data analysis (stage aim).
Check datarobot_project_status_get first: the stage must be aim.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5
- `cv_method` (`string`, optional): The partitioning method to be applied to the training data. One of: random, user, stratified, group, datetime. Example: random. One of: `random`, `user`, `stratified`, `group`, `datetime`.
- `featurelist_id` (`string`, optional): Featurelist to model on (default Informative Features). Example: 5e3c94aff86f2d10692497c1
- `holdout_pct` (`number`, optional): The percentage of the dataset to assign to the holdout set. Example: 10
- `metric` (`string`, optional): Optimization metric (default chosen by DataRobot). Example: LogLoss
- `mode` (`string`, optional): Modeling mode. Example: quick One of: 0, 2, 4, 3, auto, manual, comprehensive, quick. One of: `0`, `2`, `4`, `3`, `auto`, `manual`, `comprehensive`, `quick`.
- `positive_class` (`string`, optional): For binary classification, the class to treat as positive. Example: yes
- `quickrun` (`boolean`, optional): (deprecated): 'quick' should be used in the mode parameter instead of using this parameter. if set to true, autopilot mode will be set to 'quick'.cannot be set to true when mode is set to 'comprehensive' or 'manual'. Example: true
- `reps` (`integer`, optional): The number of cross validation folds to use. Example: 10
- `seed` (`integer`, optional): A seed to use for randomization. Example: 10
- `target` (`string`, optional): Name of the column to predict. Example: churned
- `target_type` (`string`, optional): Used to specify the targettype to use for a project when it is ambiguous, i.e. a numeric target with a few unique values that could be used for either regression or multiclass. One of: Binary, Regression, Multiclass, Multilabel. Example: Binary. One of: `Binary`, `Regression`, `Multiclass`, `Multilabel`.
- `unsupervised_mode` (`boolean`, optional): If true, unsupervised project (without target) will be created. target cannot be specified if unsupervisedmode is true. Example: false
- `use_gpu` (`boolean`, optional): Indicates whether project should use gpu workers. Example: true
- `validation_pct` (`number`, optional): The percentage of the dataset to assign to the validation set. Example: 10
- `validation_type` (`string`, optional): The validation method to be used. cv for cross validation or tvh for train-validation-holdout split. One of: CV, TVH. Example: CV. One of: `CV`, `TVH`.
- `weights` (`string`, optional): The name of a column specifying row weights. the data in this column must be pure numeric (e.g. not currency, date, length, etc.) and without missing values. Example: weight

Example input: `{"project_id":"<project_id>"}`

### `datarobot_project_update`

Update Project · Write

Rename a project, change its description or worker count, or unlock its holdout.
Returns the JSON response from the API.
Use datarobot_project_update to edit project settings.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5
- `gpu_worker_count` (`integer`, optional): Number of GPU workers; -1 uses the maximum. Example: 0
- `holdout_unlocked` (`boolean`, optional): Set to true to unlock the holdout (cannot be undone). Example: true
- `project_description` (`string`, optional): New description. Example: Retrained with Q4 data
- `project_name` (`string`, optional): New project name. Example: Customer churn v2
- `worker_count` (`integer`, optional): Number of modeling workers; -1 uses the maximum. Example: 4

Example input: `{"project_id":"<project_id>"}`

### `datarobot_use_case_create`

Create Use Case · Write

Create a Use Case.
Returns the new Use Case with its id.
Use datarobot_use_case_create to start a new workspace for related projects and datasets.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `description` (`string`, optional): Use Case description. Example: Predict which customers will cancel next quarter
- `name` (`string`, optional): Use Case name. Example: Customer churn

Example input: `{}`

### `datarobot_use_case_update`

Update Use Case · Write

Rename a Use Case or change its description.
Returns an object with advancedTour, applicationsCount, created, createdAt, createdBy, customApplicationSourcesCount, customApplicationsCount, customJobsCount, customModelVersionsCount, datasetsCount, and more.
Use datarobot_use_case_update to edit a Use Case.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `use_case_id` (`string`, required): Use Case ID, from datarobot_use_cases_list. Example: 65a1b2c3d4e5f6a7b8c9d0e1
- `description` (`string`, optional): New description. Example: Churn model for 2027 planning
- `name` (`string`, optional): New name. Example: Customer churn 2027

Example input: `{"use_case_id":"<use_case_id>"}`

### `datarobot_batch_prediction_cancel`

Cancel Batch Prediction Job · Destructive

Abort a running batch prediction job and delete it with its data.
Returns the JSON response from the API.
Use datarobot_batch_prediction_cancel to stop and remove a job.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `prediction_job_id` (`string`, required): Batch prediction job ID. Example: 5e3c94aff86f2d10692497c5

Example input: `{"prediction_job_id":"<prediction_job_id>"}`

### `datarobot_dataset_delete`

Delete Dataset · Destructive

Mark a dataset as deleted (soft delete).
Returns the JSON response from the API.
Use datarobot_dataset_delete to remove a dataset from the Data Registry.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `dataset_id` (`string`, required): Dataset ID, from datarobot_datasets_list. Example: 5e3c94aff86f2d10692497b7

Example input: `{"dataset_id":"<dataset_id>"}`

### `datarobot_deployment_delete`

Delete Deployment · Destructive

Delete a deployment.
Returns the JSON response from the API.
Use datarobot_deployment_delete only to retire a deployment for good.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `deployment_id` (`string`, required): Deployment ID, from datarobot_deployments_list. Example: 5e3c94aff86f2d10692497b6
- `ignore_management_agent` (`string`, optional): If true, do not wait for the management agent to delete the deployment first. Example: false One of: false, true. One of: `false`, `true`.

Example input: `{"deployment_id":"<deployment_id>"}`

### `datarobot_model_delete`

Delete Model · Destructive

Remove a model from a project leaderboard.
Returns the JSON response from the API.
Use datarobot_model_delete to clean up models you will not use.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `model_id` (`string`, required): Model ID, from datarobot_models_list. Example: 5e3c94aff86f2d10692497b8
- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5

Example input: `{"model_id":"<model_id>","project_id":"<project_id>"}`

### `datarobot_project_delete`

Delete Project · Destructive

Delete a project and its models.
Returns the JSON response from the API.
Use datarobot_project_delete only to remove a project for good.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5

Example input: `{"project_id":"<project_id>"}`

### `datarobot_project_job_cancel`

Cancel Project Job · Destructive

Cancel a pending modeling job.
Returns the JSON response from the API.
Use datarobot_project_job_cancel to stop a job you no longer need.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `job_id` (`string`, required): Job ID. Example: 42
- `project_id` (`string`, required): Project ID, from datarobot_projects_list. Example: 5e3c94aff86f2d10692497b5

Example input: `{"job_id":"<job_id>","project_id":"<project_id>"}`

### `datarobot_use_case_delete`

Delete Use Case · Destructive

Delete a Use Case.
Returns the JSON response from the API.
Use datarobot_use_case_delete only to remove a Use Case for good.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `use_case_id` (`string`, required): Use Case ID, from datarobot_use_cases_list. Example: 65a1b2c3d4e5f6a7b8c9d0e1

Example input: `{"use_case_id":"<use_case_id>"}`

## Related

Other AI connectors ([all 216](/agentkit/connectors/?category=ai)):

- [Chorus](/agentkit/connectors/chorus/): Scalekit connector, Username and password, 9 tools
- [Gong](/agentkit/connectors/gong/): Scalekit connector, OAuth, 69 tools
- [Jiminny](/agentkit/connectors/jiminny/): Scalekit connector, Access token, 27 tools


---

## More Scalekit documentation

| Resource | What it contains | When to use it |
|----------|-----------------|----------------|
| [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content |
| [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas |
| [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |
