Nuclia Agentic RAG

https://docs.rag.progress.cloud/docs/
Admin
Nuclia Agentic RAG is an advanced retrieval-augmented generation platform that enables semantic...
Tokens:537,948
Snippets:3,989
Trust Score:5.7
Update:1 week ago
Show doc for...
Tokens:
Raw
### Install and run NucliaDB with pip

Source: https://docs.rag.progress.cloud/docs/develop/python_nucliadb_datasets

Installs the nucliadb package via pip and starts the NucliaDB server with the `nucliadb` command.

```bash
pip install nucliadb
nucliadb
```

--------------------------------

### SuggestAsync usage example

Source: https://docs.rag.progress.cloud/docs/develop/dotnet-sdk/ISearchService

Use this example to get suggestions. It prints the count of paragraph and entity suggestions.

```csharp
var response = await searchService.SuggestAsync("authentication");  
if (response.IsSuccessful)  
{  
    var suggestions = response.Data;  
    Console.WriteLine($"Found {suggestions.Paragraphs?.Count ?? 0} paragraph suggestions");  
    Console.WriteLine($"Found {suggestions.Entities?.Count ?? 0} entity suggestions");  
}  

```

--------------------------------

### GetLearningSchemaAsync example

Source: https://docs.rag.progress.cloud/docs/develop/dotnet-sdk/IKnowledgeBoxService

Example of retrieving the learning schema and printing the number of top-level properties.

```csharp
var response = await knowledgeBoxService.GetLearningSchemaAsync();  
if (response.IsSuccessful)  
{  
    var schema = response.Data;  
    Console.WriteLine($"Schema has {schema.Count} top-level properties");  
}  
```

--------------------------------

### Run website upload script with example URL

Source: https://docs.rag.progress.cloud/docs/ingestion/how-to/index-website

Example command that uploads the Progress RAG documentation website.

```bash
python3 script.py https://docs.rag.progress.cloud/docs/  
```

--------------------------------

### DownloadKnowledgeBoxExportAsync example

Source: https://docs.rag.progress.cloud/docs/develop/dotnet-sdk/IKnowledgeBoxService

Example of downloading an export and saving it to a file.

```csharp
var response = await knowledgeBoxService.DownloadKnowledgeBoxExportAsync("my-kb-id", "export-123");  
if (response.IsSuccessful)  
{  
    await File.WriteAllBytesAsync("kb-export.zip", response.Data);  
    Console.WriteLine("Export downloaded successfully");  
}  
```

--------------------------------

### Create a task (SDK)

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/da-agents

Starts a data augmentation task using the Nuclia SDK. The example uses the LABELER task name with ApplyOptions.EXISTING and a DataAugmentation parameters object.

```python
from nuclia import sdk  
from nuclia_models.worker.tasks import TaskName, ApplyOptions, DataAugmentation  
from nuclia_models.worker.proto import ApplyTo, Filter, LLMConfig, Operation  
  
kb = sdk.NucliaKB()  
kb.task.start(  
    task_name=TaskName.LABELER,  
    apply=ApplyOptions.EXISTING,  
    parameters=DataAugmentation(  
        name="test",  
        on=ApplyTo.FIELD,  
        filter=Filter(),  
        operations=[Operation()],  
        llm=LLMConfig(),  
    ),  
)  
```

--------------------------------

### Create a task (CLI)

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/da-agents

Starts a data augmentation task using the Nuclia CLI. The example uses the labeler task name with apply=existing and a JSON parameters object.

```bash
nuclia kb task start --task_name=labeler --apply=existing --parameters='{  
    "name": "test",  
    "on": 1,  
    "filter": {},  
    "operations": [  
        {}  
    ],  
    "llm": {}  
}'  
```

--------------------------------

### Install Nuclia SDK with pip

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/README

Install the Nuclia SDK/CLI. Requires Python 3.8 or higher.

```bash
pip install nuclia  
```

--------------------------------

### GetKnowledgeBoxConfigurationAsync usage example

Source: https://docs.rag.progress.cloud/docs/develop/dotnet-sdk/IKnowledgeBoxService

C# example that calls GetKnowledgeBoxConfigurationAsync and iterates over the configuration dictionary, printing each key-value pair.

```csharp
var response = await knowledgeBoxService.GetKnowledgeBoxConfigurationAsync("my-kb-id");  
if (response.IsSuccessful)  
{  
    var config = response.Data;  
    foreach (var setting in config)  
    {  
        Console.WriteLine($"{setting.Key}: {setting.Value}");  
    }  
}  
```

--------------------------------

### Install @nuclia/core via npm

Source: https://docs.rag.progress.cloud/docs/develop/js-sdk

Install the Nuclia SDK package via npm.

```bash
npm install @nuclia/core  
```

--------------------------------

### Install the NucliaDB Python SDK

Source: https://docs.rag.progress.cloud/docs/management/nucliadb/intro

Installs the nucliadb_sdk package, required to interact with NucliaDB programmatically.

```bash
pip install nucliadb_sdk  
```

--------------------------------

### Install Nuclia Python SDK

Source: https://docs.rag.progress.cloud/docs/ingestion/how-to/google-colab

Installs the Nuclia Python SDK using pip in the Colab environment.

```bash
!pip install nuclia  
```

--------------------------------

### Get Usage - 200 Response Example (application/json)

Source: https://docs.rag.progress.cloud/docs/account-api

Example JSON response for the Get Usage endpoint (GET /api/v1/account/{account_id}/usage) with HTTP 200 status. Shows the structure with timestamp and metrics array, where metrics include name, value, and details.

```json
[
  {
    "timestamp": "2019-08-24T14:15:22Z",
    "metrics": [
      {
        "name": "pre_processing_time",
        "value": 0,
        "details": []
      }
    ]
  }
]
```

--------------------------------

### Show configured authentication using CLI

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/auth

List configured user accounts, knowledge boxes, and NUA keys.

```bash
nuclia auth show  

```

--------------------------------

### Get Authenticated User - 200 Response Example

Source: https://docs.rag.progress.cloud/docs/account-api

Example JSON response for the GET /api/v1/user endpoint. Returns user information including id, name, email, language, type, last_login, and avatar.

```json
{
  "id": "string",
  "name": "string",
  "email": "string",
  "language": "CA",
  "type": "ROOT",
  "last_login": "2019-08-24T14:15:22Z",
  "avatar": "string"
}
```

--------------------------------

### Get source response example (NucliaDBSourceSchema)

Source: https://docs.rag.progress.cloud/docs/nucliadb-agentic-api

Example response body for the GET /kb/{kbid}/sources/{source_id} endpoint. Shows the structure of a NucliaDB source with type, description, search_config, filter_expression, labels, and resource_filters.

```json
{
  "type": "nucliadb",
  "description": "string",
  "search_config": "string",
  "filter_expression": {
    "field": { },
    "paragraph": { },
    "key_value": { },
    "operator": "and"
  },
  "labels": [
    "string"
  ],
  "resource_filters": [
    "string"
  ]
}
```

--------------------------------

### Setup: imports and constants for KB management

Source: https://docs.rag.progress.cloud/docs/management/how-to/manage-kbs

Defines the required imports and constants for the script, including the management API endpoints, zone, account ID, and NUA API key. Replace the placeholder values with your own.

```python
import requests  
  
MANAGEMENT_API = "https://accounts.progress.cloud/api/v1"  
ZONE = "europe-1"  
ZONE_MANAGEMENT_API = f"https://{ZONE}.dp.progress.cloud/api/v1"  
ACCOUNT_ID = "YOUR-ACCOUNT-ID"  
  
NUA_API_KEY = "YOUR-NUA-KEY"  

```

--------------------------------

### Create a NucliaDB SDK instance

Source: https://docs.rag.progress.cloud/docs/develop/python_nucliadb_datasets

Creates a `NucliaDB` SDK instance for connecting to a Nuclia KB. The example uses `Region.EUROPE1` and a URL for rag.progress.cloud, but the same pattern works for local NucliaDB with `Region.ON_PREM`.

```python
from nucliadb_sdk import Region, NucliaDB  
  
sdk = NucliaDB(  
    api_key=my_api_key,  
    region=Region.EUROPE1,  
    url="https://europe-1.dp.progress.cloud/api",  
)  

```

--------------------------------

### start(type, params?)

Source: https://docs.rag.progress.cloud/docs/develop/js-sdk/classes/Training

Starts a training task of the specified type, with optional parameters for valid labelsets and nertags.

```APIDOC
## start(type, params?)

### Description
Starts a training task and returns an Observable of TrainingTask.

### Parameters
- **type** (TrainingType) - Required - The type of training task.
- **params** (object) - Optional - Additional parameters.
  - **params.valid_labelsets** (string[]) - Optional - Valid labelsets.
  - **params.valid_nertags** (string[]) - Optional - Valid NER tags.

### Returns
`Observable<TrainingTask>`
```

--------------------------------

### Get Account - 200 Response Example (application/json)

Source: https://docs.rag.progress.cloud/docs/account-api

Example JSON response for the Get Account endpoint (GET /api/v1/account/{account_slug_or_id}) with HTTP 200 status. Shows the full account object including id, slug, title, type, zone visibility, limits, SAML configuration, and workflow.

```json
{
  "id": "string",
  "slug": "string",
  "title": "string",
  "type": "stash-trial",
  "zone_visibility": "DEFAULT",
  "description": "string",
  "can_manage_account": true,
  "current_users": 0,
  "max_users": 0,
  "current_kbs": 0,
  "max_kbs": 0,
  "current_agents": 0,
  "max_agents": 0,
  "current_memories": 0,
  "max_memories": 0,
  "current_arags": 0,
  "max_arags": 0,
  "blocking_state": "unblocked",
  "blocked_features": [],
  "limits": {
    "upload": {
      "upload_limit_max_media_file_size": 0,
      "upload_limit_max_non_media_file_size": 0
    },
    "usage": {
      "monthly_limit_paragraphs_processed": 0,
      "monthly_limit_docs_no_media_processed": 0,
      "monthly_limit_media_seconds_processed": 0,
      "monthly_limit_paragraphs_stored": 0,
      "monthly_limit_hosted_searches_performed": 0,
      "monthly_limit_hosted_answers_generated": 0,
      "monthly_limit_self_hosted_searches_performed": 0,
      "monthly_limit_self_hosted_answers_generated": 0,
      "storage_limit_max_bytes_per_kb": 0,
      "storage_limit_max_resources_per_kb": 0,
      "monthly_limit_non_media_files_processed": 0,
      "monthly_limit_chars_processed": 0
    },
    "processing": {
      "monthly_limit_paragraphs_processed": 0,
      "monthly_limit_docs_no_media_processed": 0,
      "monthly_limit_media_seconds_processed": 0,
      "monthly_limit_paragraphs_stored": 0,
      "monthly_limit_hosted_searches_performed": 0,
      "monthly_limit_hosted_answers_generated": 0,
      "monthly_limit_self_hosted_searches_performed": 0,
      "monthly_limit_self_hosted_answers_generated": 0,
      "storage_limit_max_bytes_per_kb": 0,
      "storage_limit_max_resources_per_kb": 0,
      "monthly_limit_non_media_files_processed": 0,
      "monthly_limit_chars_processed": 0
    }
  },
  "creation_date": "2019-08-24T14:15:22Z",
  "trial_expiration_date": "2019-08-24T14:15:22Z",
  "domain": "string",
  "saml_entity_id": "string",
  "saml_sso_url": "string",
  "saml_x509_cert": "string",
  "saml_config": {
    "domains": [
      "string"
    ],
    "sso_url": "string",
    "entity_id": "string",
    "x509_cert": "string",
    "authn_context": "exact"
  },
  "workflow": "classic",
  "eula_accepted": false
}
```

--------------------------------

### GET request with min_score query params

Source: https://docs.rag.progress.cloud/docs/rag/advanced/search

An example GET request to the search endpoint with min_score parameters as query parameters.

```http
/api/v1/kb/kbid/search?query=Hakuna%20Matata&min_score_bm25=4.0&min_score_semantic=1.2  
```

--------------------------------

### Initialize Nuclia client

Source: https://docs.rag.progress.cloud/docs/ingestion/how-to/integrate-nextjs

Shows how to instantiate the Nuclia client with your backend, zone, knowledge box ID, and API key. The API key is required for adding or modifying content.

```javascript
const { Nuclia } = require("@nuclia/core");  
require("localstorage-polyfill");  
require("isomorphic-unfetch");  
  
const nuclia = new Nuclia({  
  backend: "https://accounts.progress.cloud/api",  
  zone: "europe-1",  
  knowledgeBox: "<YOUR-KB-ID>",  
  apiKey: "<YOUR-API-KEY>",  
});  
  
// code to push data to Agentic RAG (detailed later)  
```

--------------------------------

### GET query params for min_score

Source: https://docs.rag.progress.cloud/docs/rag/search-strategy

Example query parameters for GET versions of search and find endpoints to specify minimum scores.

```http
/api/v1/kb/kbid/search?query=Hakuna%20Matata&min_score_bm25=4.0&min_score_semantic=1.2  

```

--------------------------------

### CLI: List resources

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/memory

Command-line equivalents for listing resources, with optional query, page, and size filters.

```bash
nuclia memory list_resources  
nuclia memory list_resources --query=policy --page=0 --size=10  
```

--------------------------------

### List Accounts - 200 Response Example

Source: https://docs.rag.progress.cloud/docs/account-api

Example JSON response for the GET /api/v1/accounts endpoint. Returns an array of account objects with fields id, slug, title, type, and zone_visibility.

```json
[
  {
    "id": "string",
    "slug": "string",
    "title": "string",
    "type": "stash-trial",
    "zone_visibility": "DEFAULT"
  }
]
```

--------------------------------

### Start Copilot CLI

Source: https://docs.rag.progress.cloud/docs/develop/openai_compat

Start Copilot CLI. Models used with Copilot CLI must support streaming and tool calling.

```bash
copilot  
```

--------------------------------

### Get Account User Details - 200 Response

Source: https://docs.rag.progress.cloud/docs/account-api

Example JSON response for a successful GET request to retrieve a specific account user. Contains user id, name, email, and role.

```json
{
  "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  "name": "string",
  "email": "string",
  "role": "AOWNER"
}
```

--------------------------------

### CreateResourceAsync example

Source: https://docs.rag.progress.cloud/docs/develop/dotnet-sdk/IResourceService

Example of creating a resource with a title, summary, and icon, then checking the response and printing the created resource's UUID.

```csharp
var resourceData = new CreateResourcePayload  
{  
    Title = "My New Document",  
    Summary = "This is a sample document",  
    Icon = "application/pdf"  
};  
  
var response = await resourceService.CreateResourceAsync(resourceData);  
if (response.IsSuccessful)  
{  
    var created = response.Data;  
    Console.WriteLine($"Resource created with ID: {created.Uuid}");  
}  
```

--------------------------------

### JSON example for indexing structured text

Source: https://docs.rag.progress.cloud/docs/ingestion/indexing

Example JSON array that can be indexed by Agentic RAG. The structure allows asking questions like 'Who wrote The Lord of the Rings?' and getting the expected answer.

```json
[
  {
    "book": "The Lord of the Rings",
    "author": "J. R. R. Tolkien"
  },
  {
    "book": "Dune",
    "author": "Frank Herbert"
  }
]
```

--------------------------------

### getResource example

Source: https://docs.rag.progress.cloud/docs/develop/js-sdk/classes/KnowledgeBox

Retrieves a resource from the Knowledge Box by UUID using the getResource method. The example uses switchMap to get the KnowledgeBox instance and then calls getResource with a specific UUID, logging the resource to the console.

```typescript
nuclia.db  
  .getKnowledgeBox()  
  .pipe(switchMap((knowledgeBox) => knowledgeBox.getResource('09a94719a6444c5a9689394f6ed9baf6')))  
  .subscribe((resource) => {  
    console.log('resource', resource);  
  });  

```

--------------------------------

### CLI: Create a resource

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/memory

Command-line equivalent for creating a resource with a title and slug.

```bash
nuclia memory create_resource --title="Vacation Policy" --slug=vacation-policy  
```

--------------------------------

### Create personal token and login using CLI

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/auth

Generate a personal access token and use it as authentication in the CLI.

```bash
nuclia auth create_personal_token --description="My token" --days=30 --login  

```

--------------------------------

### Example request to /api/status

Source: https://docs.rag.progress.cloud/docs/management/how-to/status

Curl command to send a GET request to the `/status` endpoint. No authorization is required.

```bash
curl -X GET "https://status-api.nuclia.com/api/status"  

```

--------------------------------

### modify() example: Update resource description

Source: https://docs.rag.progress.cloud/docs/develop/js-sdk/classes/Session

This example shows how to use the modify() method to update a resource's description. It chains observables to get the knowledge box, then the resource, and calls modify with the new description. The synchronous parameter defaults to true.

```typescript
nuclia.db  
  .getKnowledgeBox('my-account', 'my-kb')  
  .pipe(  
    switchMap((knowledgeBox) => knowledgeBox.getResource('my-resource')),  
    switchMap((resource) => resource.modify({ description: 'new description' })),  
  )  
  .subscribe(() => {  
    console.log('resource modified');  
  });  
```

--------------------------------

### OpenCode authentication file

Source: https://docs.rag.progress.cloud/docs/develop/openai_compat

Example ~/.local/share/opencode/auth.json for storing the NUA key.

```json
{  
  "NUA AWS": {  
    "type": "api",  
    "key": "<NUA KEY>"  
  }  
}  
```

--------------------------------

### Retrieve a session from the Retrieval Agent using getSession()

Source: https://docs.rag.progress.cloud/docs/develop/js-sdk/classes/RetrievalAgent

Example usage of the getSession method. It retrieves a session by session ID from the Retrieval Agent. The example uses switchMap to get the agent and then calls getSession with a specific session ID, logging the session to the console.

```typescript
nuclia.db  
  .getRetrievalAgent()  
  .pipe(switchMap((agent) => agent.getSession('09a94719a6444c5a9689394f6ed9baf6')))  
  .subscribe((session) => {  
    console.log('session', session);  
  });  
```

--------------------------------

### Create Account - 201 Response Example

Source: https://docs.rag.progress.cloud/docs/account-api

Example JSON response for a successful account creation (HTTP 201). Contains the new account's id.

```json
{
  "id": "string"
}
```

--------------------------------

### new Training(kb, nuclia)

Source: https://docs.rag.progress.cloud/docs/develop/js-sdk/classes/Training

Constructor for the Training class. Initializes a new Training instance with the given knowledge box and Nuclia client.

```APIDOC
## new Training(kb, nuclia)

### Description
Creates a new Training instance.

### Parameters
- **kb** (WritableKnowledgeBox) - Required - The knowledge box to manage training for.
- **nuclia** (INuclia) - Required - The Nuclia client instance.

### Returns
`Training`
```

--------------------------------

### List Account Users - 200 Response

Source: https://docs.rag.progress.cloud/docs/account-api

Example JSON response for a successful GET request to list all users of an account. Returns an array of user objects.

```json
[
  {
    "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
    "name": "string",
    "email": "string",
    "role": "AOWNER"
  }
]
```

--------------------------------

### List resources with Nuclia CLI

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/kb

Use the `nuclia kb list` command to list resources. The `--page` and `--size` flags control pagination.

```bash
nuclia kb list  
nuclia kb list --page=0 --size=10  

```

--------------------------------

### GetResourceBySlugAsync - Get resource by slug

Source: https://docs.rag.progress.cloud/docs/develop/dotnet-sdk/IResourceService

Retrieves a resource using its human-readable slug. The example uses a placeholder slug and prints the resource's Id after checking IsSuccessful.

```csharp
Task<ApiResponse<NucliadbModelsResourceResource>> GetResourceBySlugAsync(string slug)  
```

```csharp
var response = await resourceService.GetResourceBySlugAsync("my-document-slug");  
if (response.IsSuccessful)  
{  
    var resource = response.Data;  
    Console.WriteLine($"Found resource: {resource.Id}");  
}  
```

--------------------------------

### Run nucliadb CLI with PG and S3

Source: https://docs.rag.progress.cloud/docs/management/nucliadb/deploy/basics

Starts NucliaDB using PostgreSQL as the key-value driver and S3 as the file backend. Requires a running PostgreSQL and S3-compatible service (e.g., Minio). Replace HOSTNAME, minio_server_ip, and MY_KEY with your values.

```bash
nucliadb --driver PG --file-backend S3 \
  --driver-pg-url=postgresql://postgres:password@HOSTNAME:5432/postgres \
  --s3-endpoint=http://<minio_server_ip>:9000 \
  --s3-client-id=admin \
  --s3-client-secret=12345678 \
  --s3-bucket=nucliadb-{kbid} \
  --nua-api-key="MY_KEY"  

```

--------------------------------

### Create a Backup with NucliaBackup().create()

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/kb-backup

Creates a backup of a knowledge box. Backup creation is asynchronous; the returned object represents the request, and the `finished_at` and `size` fields are populated once the backup completes.

```python
from nuclia import sdk  
from nuclia_models.accounts.backups import BackupCreate  

# Authenticate and select the account  
sdk.NucliaAuth().login()  
sdk.NucliaAccounts().default("<your-account-slug>")  

# Create a backup  
backup = sdk.NucliaBackup().create(  
    backup=BackupCreate(kb_id="<your-kb-id>"),  
    zone="<zone>"  
)  
print(backup)  
```

--------------------------------

### Set a label set

Source: https://docs.rag.progress.cloud/docs/develop/js-sdk/classes/WritableKnowledgeBox

Creates or updates a label set in the knowledge box. The example uses switchMap to get the knowledge box and then sets a label set with the ID 'status'.

```typescript
nuclia.db  
  .getKnowledgeBox('my-account', 'my-kb')  
  .pipe(  
    switchMap((knowledgeBox) =>  
      knowledgeBox.setLabelSet('status', {  
        title: 'Status',  
        color: '#ff0000',  
        labels: [{ title: 'Major' }, { title: 'Minor' }, { title: 'Critical' }],  
      }),  
    ),  
  )  
  .subscribe(() => {  
    console.log('label set set');  
  });  

```

--------------------------------

### Example response with origin metadata link

Source: https://docs.rag.progress.cloud/docs/rag/how-to/e-commerce

Example output from the /ask endpoint when using the metadata_extension strategy with origin metadata. It shows a markdown link generated from the origin URL.

```markdown
These shoes are great for dancing, [buy them now!](https://www.my-ecommerce-website.com/red-shoes)  

```

--------------------------------

### GetResourceByIdAsync - Get resource by ID

Source: https://docs.rag.progress.cloud/docs/develop/dotnet-sdk/IResourceService

Retrieves a resource by its unique identifier. The example uses a placeholder ID and prints the resource's Title and Created properties after checking IsSuccessful.

```csharp
Task<ApiResponse<NucliadbModelsResourceResource>> GetResourceByIdAsync(string resourceId)  
```

```csharp
var response = await resourceService.GetResourceByIdAsync("my-resource-id");  
if (response.IsSuccessful)  
{  
    var resource = response.Data;  
    Console.WriteLine($"Resource: {resource.Title}");  
    Console.WriteLine($"Created: {resource.Created}");  
}  
```

--------------------------------

### OpenCode configuration for AWS US region

Source: https://docs.rag.progress.cloud/docs/develop/openai_compat

Example ~/.config/opencode/opencode.json setup for the AWS US region and 3 models. You may choose any of the available models and the region that best suits your needs.

```json
{  
  "$schema": "https://opencode.ai/config.json",  
  "provider": {  
    "nuclia": {  
      "npm": "@ai-sdk/openai-compatible",  
      "name": "NUA AWS",  
      "options": {  
        "baseURL": "https://aws-us-east-2-1.dp.progress.cloud/api/v1/predict/compat"  
      },  
      "models": {  
        "aws-claude-4-5-sonnet": {  
          "name": "AWS Claude 4.5 Sonnet" ,  
          "variants": {  
            "thinking": {  
              "reasoningEffort": "high"  
            },  
            "fast": {}  
          }  
        },  
        "aws-claude-4-5-haiku": { "name": "AWS Claude 4.5 Haiku" },  
        "chatgpt4o": { "name": "GPT-4o" }  
      }  
    }  
  },  
  "model": "aws-claude-4-5-haiku",  
  "small_model": "aws-claude-4-5-sonnet"  
}  
```

--------------------------------

### Install Nuclia SDK and dependencies

Source: https://docs.rag.progress.cloud/docs/ingestion/how-to/integrate-nextjs

Install the @nuclia/core package along with localstorage-polyfill and isomorphic-unfetch to use the Nuclia API in NodeJS. Use either npm or yarn.

```bash
npm install @nuclia/core localstorage-polyfill isomorphic-unfetch  
# OR  
yarn add @nuclia/core localstorage-polyfill isomorphic-unfetch  
```

--------------------------------

### GET /api/v1/account/{account_id}/invites - 200 Successful Response

Source: https://docs.rag.progress.cloud/docs/account-api

Example JSON response for a successful list of invited users. Each item contains an email, role (e.g., AOWNER), and expiration timestamp.

```json
[
  {
    "email": "string",
    "role": "AOWNER",
    "expires": "2019-08-24T14:15:22Z"
  }
]
```

--------------------------------

### Match origin path prefix

Source: https://docs.rag.progress.cloud/docs/rag/advanced/search-filters

Matches any path starting with the provided value. For example, `Users/JohnDoe/Documents` will match files in the `Documents` folder of the `JohnDoe` user, but also the ones in `Documents/Work` or `Documents/Personal`, etc.

```json
{
  "field": {
    "prop": "origin_path",
    "prefix": "Users/JohnDoe/Documents"
  }
}
```

--------------------------------

### Install Progress.Nuclia via .NET CLI

Source: https://docs.rag.progress.cloud/docs/develop/dotnet-sdk

Adds the Progress.Nuclia NuGet package to a .NET project using the dotnet CLI.

```bash
dotnet add package Progress.Nuclia  
```

--------------------------------

### Match origin path prefix

Source: https://docs.rag.progress.cloud/docs/rag/advanced/search-filters

Matches path of the resource in the source system. It will match any path starting with the provided value. Example: `Users/JohnDoe/Documents` will match files in the `Documents` folder of the `JohnDoe` user, but also the ones in `Documents/Work` or `Documents/Personal`, etc.

```json
{  
  "field": {  
    "prop": "origin_path",  
    "prefix": "Users/JohnDoe/Documents"  
  }  
}  
```

--------------------------------

### Faceted parameter examples

Source: https://docs.rag.progress.cloud/docs/rag/search-strategy

Use the faceted parameter to get the total amount of matches for each image file type (like jpg, png, gif, etc.) and for each language (like en, it, fr, etc.). The parameter takes the same values as the filters parameter.

```text
faceted=/icon/image  
```

```text
faceted=/metadata.language  
```

--------------------------------

### Export a KB using the CLI

Source: https://docs.rag.progress.cloud/docs/develop/python-sdk/import-export

Set the default KB to the source KB, then start an export to a specified path. The command outputs an export ID and progress, then downloads the export file.

```bash
nuclia kbs default my-source-kb  
nuclia kb exports start --path=/some/path/kb.export  
  
> Export for KB started. export_id=77ae82a2acea4c708cd775e6ce0ab30b  
> Generating export. Status: scheduled: 0resources [00:03, ?resources/s]  
> Export is ready. Will be downloaded to /some/path/kb.export.  
> Downloading data: 100%|█████████████████████████████████████████████████████████| 3.66M/3.66M [00:00<00:00, 14.1MiB/s  
```

--------------------------------

### GET /agent/{agent_id}/prompts - 200 Response Sample

Source: https://docs.rag.progress.cloud/docs/hyperforge-api

Example JSON response for a successful request to retrieve agent prompts. The response is an array of prompt objects, each containing name, description, prompt, arguments, icons, meta, and prompt_id fields.

```json
[
  {
    "name": "string",
    "description": "string",
    "prompt": "string",
    "arguments": [
      {
        "name": "string",
        "description": "string",
        "required": true
      }
    ],
    "icons": {
      "property1": "string",
      "property2": "string"
    },
    "meta": {
      "property1": "string",
      "property2": "string"
    },
    "prompt_id": "string"
  }
]
```

--------------------------------

### Docker Compose setup for NucliaDB with Basic Auth (auth_policy=upstream_basicauth)

Source: https://docs.rag.progress.cloud/docs/management/nucliadb/deploy/access

Configures NucliaDB and PostgreSQL using Docker Compose for internal API access with Basic Auth. Set the `auth_policy` environment variable to `upstream_basicauth` and provide your NUA API key. The example includes CORS origins and a PostgreSQL database.

```yaml
version: '3.9'  
  
services:  
  nucliadb:  
    image: nuclia/nucliadb:latest  
    ports:  
      - '8080:8080'  
    environment:  
      - NUA_API_KEY=<my-key>  
      - NUCLIA_ZONE=europe-1  
      - DRIVER=PG  
      - DRIVER_PG_URL=postgres://nucliadb:nucliadb@postgres:5432/nucliadb  
      - auth_policy=upstream_basicauth  
      - CORS_ORIGINS=["http://localhost:8080"]  
    volumes:  
      - nucliadb-data:/data  
    depends_on:  
      - postgres  
  
  postgres:  
    image: postgres:latest  
    ports:  
      - '5432:5432'  
    environment:  
      - POSTGRES_USER=nucliadb  
      - POSTGRES_PASSWORD=nucliadb  
      - POSTGRES_DB=nucliadb  
    volumes:  
      - nucliadb-maindb:/var/lib/postgresql/data  
  
volumes:  
  nucliadb-data: {}  
  nucliadb-maindb: {}  

```
Tokens:
Raw