> **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 Nanonets

The Nanonets connector lets your AI agent act in each user's Nanonets account. Each user connects their own Nanonets API key once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 28 tools.

**Authentication:** API Key
**Categories:** AI, Files & Documents
**Tools:** 28: 11 read, 15 write, 2 destructive
**Users sign in with:** API key
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=NANONETS)

## 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 Nanonets connection

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

## 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: 'nanonets_classification_model_list',
  toolInput: {},
  connector: 'nanonets',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="nanonets_classification_model_list",
    tool_input={},
    connection_name="nanonets",
    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.

### `nanonets_classification_model_list`

List Image Classification Models · Read-only

List the image classification models in the Nanonets account using the v2 API.
Returns each model's model_id, model_type, state and categories with counts.
Use nanonets_classification_model_list to find classification models. Use nanonets_workflow_list for OCR and extraction workflows.

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

Inputs: none.

Example input: `{}`

### `nanonets_document_get`

Get Document · Read-only

Get the processing status and extracted data of one document in a Nanonets workflow.
Returns status, metadata and verification status, and for each page its page_id with extracted fields (field_data_id, value, confidence, bbox) and tables (table_id, cells).
Use nanonets_document_get to read results, including after an async upload. Use nanonets_document_page_get for a single page of a long document.

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

Inputs:

- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"document_id":"<document_id>","workflow_id":"<workflow_id>"}`

### `nanonets_document_list`

List Documents · Read-only

List the documents processed by a Nanonets workflow, newest first, one page at a time.
Returns document_id, status, uploaded_at, metadata, original_document_name, verification status and page ids, plus total_count.
Use nanonets_document_list to browse or poll uploads. Use nanonets_document_get for the extracted fields and tables of one document.

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

Inputs:

- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.
- `limit` (`integer`, optional): Number of documents per page, from 1 to 100. Defaults to 10. Example: 50. Default: `10`.
- `page` (`integer`, optional): Page number of the results, starting at 1. Defaults to 1. Example: 2. Default: `1`.

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

### `nanonets_document_page_get`

Get Document Page · Read-only

Get the extracted data of one page of a document in a Nanonets workflow.
Returns the page_id, page_number, image_url and the page's extracted fields and tables with their ids.
Use nanonets_document_page_get for page-level review of a long document. Use nanonets_document_get for all pages.

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

Inputs:

- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `page_id` (`string`, required): ID of a page of the document. Returned as pages[].page_id by nanonets_document_get. Example: b94d9ca6-37b9-11f0-a23c-367dda7a627d.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"document_id":"<document_id>","page_id":"<page_id>","workflow_id":"<workflow_id>"}`

### `nanonets_ocr_model_get`

Get OCR Model · Read-only

Get one Nanonets OCR model by ID using the v2 API.
Returns model_id, model_type, state and the model's categories (field labels) with their counts.
Use nanonets_ocr_model_get to check a v2 model's labels and training state. Use nanonets_workflow_get for the v4 workflow view.

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

Inputs:

- `model_id` (`string`, required): ID of the Nanonets OCR model (the v2 name for a workflow). Copy it from the model page in the Nanonets app, or use a workflow ID from nanonets_workflow_list. Example: 7c3a5e1f-2b4d-4f6a-8c9e-0a1b2c3d4e5f.

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

### `nanonets_ocr_prediction_file_get`

Get OCR Prediction by File · Read-only

Get the prediction results for one file processed by a Nanonets OCR model, using the v2 API.
Returns each page's predicted and moderated boxes (label, ocr_text, score, coordinates, table cells), status and request_metadata.
Use nanonets_ocr_prediction_file_get to poll an async prediction or read a whole file. Use nanonets_ocr_prediction_page_get for one page.

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

Inputs:

- `model_id` (`string`, required): ID of the Nanonets OCR model (the v2 name for a workflow). Copy it from the model page in the Nanonets app, or use a workflow ID from nanonets_workflow_list. Example: 7c3a5e1f-2b4d-4f6a-8c9e-0a1b2c3d4e5f.
- `request_file_id` (`string`, required): ID of the processed file. Returned as request_file_id by nanonets_ocr_prediction_create or nanonets_ocr_prediction_async_create. Example: 9f1c2d3e-4b5a-6c7d-8e9f-0a1b2c3d4e5f.

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

### `nanonets_ocr_prediction_list`

List OCR Predictions · Read-only

List the files a Nanonets OCR model processed within a range of days, using the v2 API.
Returns moderated_images and unmoderated_images with page id, file_url, predicted_boxes and day_since_epoch, plus their counts.
Use nanonets_ocr_prediction_list to find recent results by date. Use nanonets_ocr_prediction_file_get when you have a request_file_id.

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

Inputs:

- `current_batch_day` (`integer`, required): End of the period as the number of days since 1 January 1970 (UTC). For example 20772 is 15 November 2026.
- `model_id` (`string`, required): ID of the Nanonets OCR model (the v2 name for a workflow). Copy it from the model page in the Nanonets app, or use a workflow ID from nanonets_workflow_list. Example: 7c3a5e1f-2b4d-4f6a-8c9e-0a1b2c3d4e5f.
- `start_day_interval` (`integer`, required): Start of the period as the number of days since 1 January 1970 (UTC). For example 20765 is 8 November 2026; compute it as floor(unix_seconds / 86400).

Example input: `{"current_batch_day":1,"model_id":"<model_id>","start_day_interval":1}`

### `nanonets_ocr_prediction_page_get`

Get OCR Prediction by Page · Read-only

Get the prediction results for one page processed by a Nanonets OCR model, using the v2 API.
Returns the page's initial prediction, any moderated values and the final result, with labels, ocr_text, scores and table cells.
Use nanonets_ocr_prediction_page_get for a single page. Use nanonets_ocr_prediction_file_get for every page of a file.

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

Inputs:

- `model_id` (`string`, required): ID of the Nanonets OCR model (the v2 name for a workflow). Copy it from the model page in the Nanonets app, or use a workflow ID from nanonets_workflow_list. Example: 7c3a5e1f-2b4d-4f6a-8c9e-0a1b2c3d4e5f.
- `page_id` (`string`, required): ID of the processed page. Returned as id in each result of nanonets_ocr_prediction_create, or in nanonets_ocr_prediction_list. Example: 75ff0d9d-066b-11f0-b31d-62e85b8f4aab.

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

### `nanonets_workflow_get`

Get Workflow · Read-only

Get one Nanonets workflow by ID.
Returns the workflow's description, fields and table headers (with their ids), settings and created_at.
Use nanonets_workflow_get to read a known workflow's field configuration. Use nanonets_workflow_list to browse.

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

Inputs:

- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

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

### `nanonets_workflow_list`

List Workflows · Read-only

List all document extraction workflows in the Nanonets account.
Returns each workflow's id, description, configured fields, table headers, settings and created_at.
Use nanonets_workflow_list to find a workflow_id. Use nanonets_workflow_get for one workflow you already know.

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

Inputs: none.

Example input: `{}`

### `nanonets_workflow_type_list`

List Workflow Types · Read-only

List the workflow types that can be used to create a Nanonets workflow.
Returns each type's id (for example invoice, receipt, passport, or an empty string for instant learning), name and description.
Use nanonets_workflow_type_list before nanonets_workflow_create to choose a workflow_type.

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

Inputs: none.

Example input: `{}`

### `nanonets_document_field_add`

Add Document Field Value · Write

Add a field value that extraction missed to one page of a document in a Nanonets workflow.
Returns the new field_data_id, field_name and value.
Use nanonets_document_field_add when a field has no extracted value. Use nanonets_document_field_update to correct an existing value.

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

Inputs:

- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `field_name` (`string`, required): Name of a field configured on the workflow (see nanonets_workflow_get). Example: invoice_number.
- `page_id` (`string`, required): ID of a page of the document. Returned as pages[].page_id by nanonets_document_get. Example: b94d9ca6-37b9-11f0-a23c-367dda7a627d.
- `value` (`string`, required): The value to record for the field. Example: INV-2024-004.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"document_id":"<document_id>","field_name":"<field_name>","page_id":"<page_id>","value":"<value>","workflow_id":"<workflow_id>"}`

### `nanonets_document_field_update`

Update Document Field Value · Write

Correct one extracted field value on a document page in a Nanonets workflow.
Returns the field_data_id, field_name and new value; Nanonets keeps the original value in its moderation history.
Use nanonets_document_field_update to fix a wrong value. Use nanonets_document_field_add when the field has no value yet.

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

Inputs:

- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `field_data_id` (`string`, required): ID of one extracted field value on the page. Returned as field_data_id inside pages[].data.fields by nanonets_document_get or nanonets_document_page_get. Example: f1a2b3c4-d5e6-4f7a-8b9c-d0e1f2a3b4c5.
- `page_id` (`string`, required): ID of a page of the document. Returned as pages[].page_id by nanonets_document_get. Example: b94d9ca6-37b9-11f0-a23c-367dda7a627d.
- `value` (`string`, required): The corrected value. Example: INV-2024-003.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"document_id":"<document_id>","field_data_id":"<field_data_id>","page_id":"<page_id>","value":"<value>","workflow_id":"<workflow_id>"}`

### `nanonets_document_field_verify`

Verify Document Field · Write

Mark one extracted field value on a document page in a Nanonets workflow as verified.
Returns the field_data_id, field_name and value.
Use nanonets_document_field_verify to approve a single value. Use nanonets_document_verify to verify the whole document.

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

Inputs:

- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `field_data_id` (`string`, required): ID of one extracted field value on the page. Returned as field_data_id inside pages[].data.fields by nanonets_document_get or nanonets_document_page_get. Example: f1a2b3c4-d5e6-4f7a-8b9c-d0e1f2a3b4c5.
- `page_id` (`string`, required): ID of a page of the document. Returned as pages[].page_id by nanonets_document_get. Example: b94d9ca6-37b9-11f0-a23c-367dda7a627d.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"document_id":"<document_id>","field_data_id":"<field_data_id>","page_id":"<page_id>","workflow_id":"<workflow_id>"}`

### `nanonets_document_table_cell_update`

Update Document Table Cell · Write

Correct the value of one cell in an extracted table on a document page in a Nanonets workflow.
Returns the cell_id, row, col, header and new text.
Use nanonets_document_table_cell_update to fix a line-item value. Use nanonets_document_field_update for a header field.

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

Inputs:

- `cell_id` (`string`, required): ID of the table cell. Returned as cells[].cell_id inside the table by nanonets_document_get or nanonets_document_page_get. Example: 1b5d3df7-3df7-420a-a82b-29dbdfd3e1b1.
- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `page_id` (`string`, required): ID of a page of the document. Returned as pages[].page_id by nanonets_document_get. Example: b94d9ca6-37b9-11f0-a23c-367dda7a627d.
- `table_id` (`string`, required): ID of an extracted table on the page. Returned as table_id inside pages[].data.tables by nanonets_document_get or nanonets_document_page_get. Example: d8e5c1d2-4e71-4d0e-babc-a845f2de4f1b.
- `value` (`string`, required): The corrected cell text. Example: Product B.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"cell_id":"<cell_id>","document_id":"<document_id>","page_id":"<page_id>","table_id":"<table_id>","value":"<value>","workflow_id":"<workflow_id>"}`

### `nanonets_document_table_verify`

Verify Document Table · Write

Mark every cell of one extracted table on a document page in a Nanonets workflow as verified.
Returns the table_id, headers and cells.
Use nanonets_document_table_verify after reviewing a table. Use nanonets_document_verify to verify the whole document.

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

Inputs:

- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `page_id` (`string`, required): ID of a page of the document. Returned as pages[].page_id by nanonets_document_get. Example: b94d9ca6-37b9-11f0-a23c-367dda7a627d.
- `table_id` (`string`, required): ID of an extracted table on the page. Returned as table_id inside pages[].data.tables by nanonets_document_get or nanonets_document_page_get. Example: d8e5c1d2-4e71-4d0e-babc-a845f2de4f1b.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"document_id":"<document_id>","page_id":"<page_id>","table_id":"<table_id>","workflow_id":"<workflow_id>"}`

### `nanonets_document_upload`

Upload Document from URL · Write

Send a document by public URL to a Nanonets workflow for data extraction.
Returns the document_id and status; in sync mode also the extracted fields and tables for each page.
Use nanonets_document_upload to process a new document. Use nanonets_document_get to fetch results of an async upload.
Documents over 3 pages are always processed async. Local file uploads are not supported by this tool; host the file at a URL first.

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

Inputs:

- `document_url` (`string`, required): Publicly reachable URL of the document (PDF, JPG, PNG, TIFF, XLS or XLSX, up to 20 MB and 500 pages). Example: https://example.com/invoices/inv-2024-001.pdf.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.
- `async` (`boolean`, optional): Process in the background and return immediately with status pending. Defaults to false (wait for results). Documents over 3 pages are processed async regardless. Example: true. Default: `false`.
- `metadata` (`string`, optional): Optional text or JSON string stored with the document and returned with its results, for example your own reference ID. Example: {"customer_id": "12345", "department": "finance"}.

Example input: `{"document_url":"<document_url>","workflow_id":"<workflow_id>"}`

### `nanonets_document_verify`

Verify Document · Write

Mark every field and table of a document in a Nanonets workflow as verified.
Returns the document_id, its new status and the verified fields and tables per page.
Use nanonets_document_verify after reviewing a whole document. Use nanonets_document_field_verify or nanonets_document_table_verify to verify one item.

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

Inputs:

- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"document_id":"<document_id>","workflow_id":"<workflow_id>"}`

### `nanonets_ocr_file_approve`

Approve OCR File · Write

Approve a processed file in a Nanonets OCR model so its review status becomes approved, using the v2 API.
Returns the Nanonets confirmation response.
Use nanonets_ocr_file_approve to sign off a file after review. Use nanonets_document_verify for documents in the v4 workflow API.
Only files that passed validation or sit in an approval stage can be approved; rejected files cannot.

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

Inputs:

- `model_id` (`string`, required): ID of the Nanonets OCR model (the v2 name for a workflow). Copy it from the model page in the Nanonets app, or use a workflow ID from nanonets_workflow_list. Example: 7c3a5e1f-2b4d-4f6a-8c9e-0a1b2c3d4e5f.
- `request_file_id` (`string`, required): ID of the file to approve. Returned as request_file_id by the prediction tools, or shown on the model's extract data page. Example: 9f1c2d3e-4b5a-6c7d-8e9f-0a1b2c3d4e5f.

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

### `nanonets_ocr_prediction_async_create`

Predict from URL (Async) · Write

Queue a file URL for background processing by a Nanonets OCR model, using the v2 API.
Returns the request_file_id and page ids with status pending; the predictions are empty until processing finishes.
Use nanonets_ocr_prediction_async_create for files over 3 pages. Then poll nanonets_ocr_prediction_file_get with the request_file_id.

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

Inputs:

- `model_id` (`string`, required): ID of the Nanonets OCR model (the v2 name for a workflow). Copy it from the model page in the Nanonets app, or use a workflow ID from nanonets_workflow_list. Example: 7c3a5e1f-2b4d-4f6a-8c9e-0a1b2c3d4e5f.
- `url` (`string`, required): Publicly reachable URL of the image or PDF to process. One URL per call. Example: https://nanonets.s3-us-west-2.amazonaws.com/test-images/test1.jpg.
- `request_metadata` (`string`, optional): Optional text stored with the file and returned in its results, for example your own reference ID. Example: order-48213.

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

### `nanonets_ocr_prediction_create`

Predict from URL (Sync) · Write

Run a Nanonets OCR model on a file URL and wait for the extracted data, using the v2 API.
Returns one result per page with the predicted labels, ocr_text, confidence score, bounding boxes, table cells, request_file_id and page id.
Use nanonets_ocr_prediction_create for files of 3 pages or fewer. Use nanonets_ocr_prediction_async_create for longer files, or nanonets_document_upload for the v4 workflow API.

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

Inputs:

- `model_id` (`string`, required): ID of the Nanonets OCR model (the v2 name for a workflow). Copy it from the model page in the Nanonets app, or use a workflow ID from nanonets_workflow_list. Example: 7c3a5e1f-2b4d-4f6a-8c9e-0a1b2c3d4e5f.
- `url` (`string`, required): Publicly reachable URL of the image or PDF to process. One URL per call. Example: https://nanonets.s3-us-west-2.amazonaws.com/test-images/test1.jpg.
- `pages_to_process` (`string`, optional): Comma-separated page numbers to process from the file. Leave empty to process every page. Example: 1,2.
- `request_metadata` (`string`, optional): Optional text stored with the file and returned in its results, for example your own reference ID. Example: order-48213.

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

### `nanonets_workflow_create`

Create Workflow · Write

Create a new Nanonets document extraction workflow, either instant learning or a pre-trained type.
Returns the new workflow with its id, description, pre-configured fields and table headers, settings and created_at.
Use nanonets_workflow_create to start extracting a new kind of document. Use nanonets_workflow_update to rename an existing one.
An instant learning workflow (empty workflow_type) needs nanonets_workflow_fields_set before documents are uploaded. An account can hold at most 20 workflows.

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

Inputs:

- `description` (`string`, required): Human-readable description of what the workflow extracts. Example: Extract data from supplier invoices.
- `workflow_type` (`string`, optional): Workflow type ID from nanonets_workflow_type_list. Use an empty string (the default) for an instant learning workflow whose fields you configure yourself, or a pre-trained type: invoice, receipt, tables, purchase_order, passport, bill_of_lading. Example: invoice. One of: ``, `invoice`, `receipt`, `tables`, `purchase_order`, `passport`, `bill_of_lading`. Default: ``.

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

### `nanonets_workflow_field_update`

Update Workflow Field · Write

Rename one field or table header of a Nanonets workflow.
Returns the field id and its new name.
Use nanonets_workflow_field_update for a targeted rename that keeps the other fields. Use nanonets_workflow_fields_set to replace the whole list.

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

Inputs:

- `field_id` (`string`, required): ID of the field or table header to rename. Returned as fields[].id or table_headers[].id by nanonets_workflow_get. Example: 550e8400-e29b-41d4-a716-446655440001.
- `name` (`string`, required): New name. Only letters, numbers and underscores, unique within the workflow. Example: invoice_number.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"field_id":"<field_id>","name":"<name>","workflow_id":"<workflow_id>"}`

### `nanonets_workflow_fields_set`

Set Workflow Fields · Write

Replace all fields and table headers that a Nanonets workflow extracts.
Returns the workflow id with the new fields and table headers, each with its id.
Use nanonets_workflow_fields_set to configure an instant learning workflow or redefine every field at once. Use nanonets_workflow_field_update to rename one field.
Any field or table header not in the request is removed. Names may contain only letters, numbers and underscores.

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

Inputs:

- `field_names` (`array`, required): Names of the header fields to extract, as a list of strings. Each name must be unique and contain only letters, numbers and underscores (at most 30 fields per workflow). Example: ["invoice_number", "total_amount", "invoice_date"].
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.
- `table_header_names` (`array`, optional): Names of the table columns to extract, as a list of strings. Same naming rules as field_names (at most 20 per workflow). Leave empty to extract no tables. Example: ["item_description", "quantity", "unit_price", "total"].

Example input: `{"field_names":["invoice_number","total_amount","invoice_date"],"workflow_id":"<workflow_id>"}`

### `nanonets_workflow_settings_update`

Update Workflow Settings · Write

Turn table capture on or off for a Nanonets workflow.
Returns the workflow id, type and updated settings.
Use nanonets_workflow_settings_update to change how new documents are processed. Use nanonets_workflow_update to change the description.
The change applies to documents uploaded after the call.

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

Inputs:

- `table_capture` (`boolean`, required): Whether the workflow extracts tables (line items) from documents. true enables table capture, false disables it. Example: true.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"table_capture":true,"workflow_id":"<workflow_id>"}`

### `nanonets_workflow_update`

Update Workflow · Write

Update the description of an existing Nanonets workflow.
Returns the workflow id, the new description and created_at.
Use nanonets_workflow_update to rename a workflow. Use nanonets_workflow_settings_update for table capture and nanonets_workflow_fields_set for fields.

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

Inputs:

- `description` (`string`, required): New description for the workflow. Example: Extract data from EU supplier invoices.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"description":"<description>","workflow_id":"<workflow_id>"}`

### `nanonets_document_delete`

Delete Document · Destructive

Permanently delete a processed document and its extracted data from a Nanonets workflow.
Returns a confirmation message.
Use nanonets_document_delete for data cleanup or retention. This cannot be undone.

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

Inputs:

- `document_id` (`string`, required): ID of the document in the workflow. Returned as document_id by nanonets_document_upload and nanonets_document_list. Example: db237bf3-f3c1-4441-905e-a6b7538db269.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"document_id":"<document_id>","workflow_id":"<workflow_id>"}`

### `nanonets_workflow_field_delete`

Delete Workflow Field · Destructive

Remove one field or table header from a Nanonets workflow.
Returns a confirmation message.
Use nanonets_workflow_field_delete to stop extracting one field. Use nanonets_workflow_fields_set to replace all fields.
This cannot be undone.

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

Inputs:

- `field_id` (`string`, required): ID of the field or table header to delete. Returned as fields[].id or table_headers[].id by nanonets_workflow_get. Example: 550e8400-e29b-41d4-a716-446655440001.
- `workflow_id` (`string`, required): ID of the Nanonets workflow. Get it from nanonets_workflow_list or nanonets_workflow_create. Example: 550e8400-e29b-41d4-a716-446655440000.

Example input: `{"field_id":"<field_id>","workflow_id":"<workflow_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 |
