### Quickstart OpenRAG Setup

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Verify prerequisites, install dependencies, and start OpenRAG with these three commands.

```bash
make check_tools
make setup
make dev
```

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

### Example User Instruction for Interactive Installation

Source: https://github.com/langflow-ai/openrag/blob/main/plugins/openrag/skills/install/SKILL.md

This is an example of how to instruct a user to perform the interactive installation steps in a separate terminal.

```text
The next step requires running an interactive command that I cannot control directly.

Please open a NEW terminal window (separate from this agent session) and run:

    uvx --python 3.13 openrag

You will see a menu. Follow these steps:
1. Type '2' and press Enter (Reconfigure)
2. Enter a STRONG OpenSearch Admin password (write it down!)
3. Enter Langflow Admin password (or autogenerate)
4. Press 'N' or skip for other prompts
5. Type '1' and press Enter (Start services)
6. Wait for "Services are running" message

Once complete, come back here and confirm: "Configuration and services started successfully"
```

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

### Install Dependencies and Start Dev Server

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/support/contribute.mdx

Install project dependencies and start a local Docusaurus static site with hot reload. The documentation will be served at `http://localhost:3000`.

```bash
npm install
npm start
```

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

### Python SDK Quick Start

Source: https://github.com/langflow-ai/openrag/blob/main/README.md

A quick example demonstrating how to use the OpenRAG Python SDK to send a message and receive a response. Ensure you have the SDK installed and an asyncio event loop running.

```python
import asyncio
from openrag_sdk import OpenRAGClient


async def main():
    async with OpenRAGClient() as client:
        response = await client.chat.create(message="What is RAG?")
        print(response.response)


if __name__ == "__main__":
    asyncio.run(main())
```

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

### Install and Run OpenRAG via uvx

Source: https://context7.com/langflow-ai/openrag/llms.txt

Quickstart installation using uvx, which automatically handles Docker dependencies. Follow the TUI prompts to configure passwords and model providers.

```bash
mkdir openrag-workspace && cd openrag-workspace
uvx --python 3.13 openrag
# Follow TUI prompts: set OpenSearch/Langflow passwords, choose model provider
```

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

### Start Podman Machine

Source: https://github.com/langflow-ai/openrag/blob/main/kubernetes/operator/README.md

Starts the Podman machine. Ensure Podman is installed and configured.

```bash
podman machine start
```

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

### Manual Infrastructure Setup

Source: https://github.com/langflow-ai/openrag/blob/main/frontend/tests/README.md

Manually starts the services for E2E tests if not using the automated script. Ensure your root .env is configured.

```bash
make dev-cpu
```

```bash
make docling
```

```bash
make health
```

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

### TypeScript/JavaScript SDK Quick Start

Source: https://github.com/langflow-ai/openrag/blob/main/README.md

A quick example demonstrating how to use the OpenRAG TypeScript/JavaScript SDK to send a message and receive a response. Ensure you have the SDK installed.

```typescript
import { OpenRAGClient } from "openrag-sdk";

const client = new OpenRAGClient();
const response = await client.chat.create({ message: "What is RAG?" });
console.log(response.response);
```

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

### Install OpenRAG MCP with uvx

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/mcp/README.md

Run the server using uvx for quick setup. Ensure Python 3.10+ and uv are installed.

```bash
uvx openrag-mcp
```

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

### Setup E2E Infrastructure Script

Source: https://github.com/langflow-ai/openrag/blob/main/frontend/tests/README.md

Automates the setup of the full stack required for E2E tests. It starts services and waits for them to be healthy.

```bash
./scripts/setup-e2e.sh
```

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

### Install Project Dependencies

Source: https://github.com/langflow-ai/openrag/blob/main/docs/README.md

Run this command to install all necessary project dependencies.

```bash
npm install
```

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

### Copy Environment Example

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/docker.mdx

Copy the example environment file to create a new .env file for your configuration.

```bash
cp .env.example .env
```

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

### Install Dependencies

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/docker.mdx

Install the necessary project dependencies using the uv sync command.

```bash
uv sync
```

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

### Clone and Setup OpenRAG Project

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Clone the OpenRAG repository and install project dependencies.

```bash
git clone https://github.com/langflow-ai/openrag.git
cd openrag
make setup
```

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

### Start Local Development Server

Source: https://github.com/langflow-ai/openrag/blob/main/docs/README.md

Starts a local development server for live previewing changes. Changes are reflected live without requiring a server restart.

```bash
npm start
```

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

### Start OpenRAG Development Environment

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Choose between starting OpenRAG with GPU support or CPU only.

```bash
make dev      # With GPU support
# or
make dev-cpu  # CPU only
```

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

### Onboarding

Source: https://context7.com/langflow-ai/openrag/llms.txt

Completes the initial OpenRAG setup by configuring models and optionally ingesting sample data.

```APIDOC
## POST /settings/onboard

### Description
Configures the initial OpenRAG setup, including selecting LLM and embedding models, and optionally ingesting sample documentation.

### Method
POST

### Endpoint
/settings/onboard

### Request Body
- **llm_provider** (string) - Required - The provider for the language model (e.g., `openai`).
- **llm_model** (string) - Required - The specific language model to use (e.g., `gpt-4o-mini`).
- **embedding_provider** (string) - Required - The provider for the embedding model (e.g., `openai`).
- **embedding_model** (string) - Required - The specific embedding model to use (e.g., `text-embedding-3-small`).
- **openai_api_key** (string) - Required if `llm_provider` is `openai` - Your OpenAI API key.

### Response Example (Success)
```json
{
  "message": "Onboarding configuration updated successfully",
  "edited": true,
  "sample_data_ingested": true,
  "task_id": "task-uuid"
}
```
```

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

### Install Python SDK

Source: https://github.com/langflow-ai/openrag/blob/main/plugins/openrag/skills/sdk/SKILL.md

Install the OpenRAG Python SDK using pip or uv.

```bash
pip install openrag-sdk
```

```bash
uv add openrag-sdk
```

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

### Install OpenRAG with uvx

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/quickstart.mdx

Install OpenRAG and its dependencies using uvx, which manages environment setup and installs Docker or Podman if needed. This command specifies Python 3.13.

```bash
uvx --python 3.13 openrag
```

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

### Install NVIDIA Container Toolkit

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/support/troubleshoot.mdx

Install the NVIDIA Container Toolkit, which includes `libnvidia-container`, `nvidia-container-toolkit-base`, and `nvidia-container-toolkit`. Version 1.14 or later is recommended for full compatibility.

```bash
sudo dnf5 install -y nvidia-container-toolkit
```

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

### Install WSL with Ubuntu

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/install-windows.mdx

Installs the Windows Subsystem for Linux with the Ubuntu distribution using WSL 2. This is the default for new installations.

```powershell
wsl --install -d Ubuntu
```

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

### Start OpenRAG Terminal Session with UVX

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/tui.mdx

Use this command to start the OpenRAG terminal session when installed with uvx.

```bash
uvx openrag
```

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

### Install OpenRAG TypeScript SDK

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/typescript/README.md

Install the SDK using npm, yarn, or pnpm.

```bash
npm install openrag-sdk
# or
yarn add openrag-sdk
# or
pnpm add openrag-sdk
```

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

### Start Docling Serve

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/docker.mdx

Starts the Docling serve process on port 5001. This is a required system-level service and must run on the host machine, not within a Docker container.

```bash
uv run python scripts/docling_ctl.py start --port 5001
```

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

### Settings Management

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/typescript/README.md

Examples for retrieving and updating SDK settings.

```APIDOC
## Settings

### Get Settings

Retrieves the current SDK settings.

```typescript
const settings = await client.settings.get();
console.log(`LLM Provider: ${settings.agent.llm_provider}`);
console.log(`LLM Model: ${settings.agent.llm_model}`);
console.log(`Embedding Model: ${settings.knowledge.embedding_model}`);
```

### Update Settings

Updates the SDK settings. Accepts an object with fields like `chunk_size` and `chunk_overlap`.

```typescript
await client.settings.update({
  chunk_size: 1000,
  chunk_overlap: 200,
});
```
```

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

### Install MCP Server Package

Source: https://github.com/langflow-ai/openrag/blob/main/plugins/openrag/skills/sdk/SKILL.md

Install the openrag-mcp package for Model Context Protocol integration using pip or uvx.

```bash
pip install openrag-mcp
```

```bash
uvx openrag-mcp
```

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

### Install Python Certificates on macOS

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/support/troubleshoot.mdx

Run this command on macOS if OpenRAG installation fails with an 'unable to get local issuer certificate' error.

```bash
open "/Applications/Python 3.13/Install Certificates.command"
```

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

### Configure MCP Server for Cursor/Claude Desktop

Source: https://github.com/langflow-ai/openrag/blob/main/README.md

Example JSON configuration for the MCP server. This setup allows AI assistants like Cursor and Claude Desktop to connect to your OpenRAG knowledge base.

```json
{
  "mcpServers": {
    "openrag": {
      "command": "uvx",
      "args": ["openrag-mcp"],
      "env": {
        "OPENRAG_URL": "http://localhost:3000",
        "OPENRAG_API_KEY": "your_api_key_here"
      }
    }
  }
}
```

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

### Install TypeScript/JavaScript SDK

Source: https://github.com/langflow-ai/openrag/blob/main/plugins/openrag/skills/sdk/SKILL.md

Install the OpenRAG SDK for TypeScript/JavaScript using npm, yarn, pnpm, or bun.

```bash
npm install openrag-sdk
```

```bash
yarn add openrag-sdk
```

```bash
pnpm add openrag-sdk
```

```bash
bun add openrag-sdk
```

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

### Install OpenRAG TypeScript/JavaScript Package

Source: https://github.com/langflow-ai/openrag/blob/main/README.md

Install the OpenRAG SDK for TypeScript/JavaScript using npm. Refer to the documentation for detailed installation options.

```bash
npm install openrag-sdk
```

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

### Verify Prerequisites with Make

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Run this command to ensure all necessary tools for OpenRAG development are installed.

```bash
make check_tools
```

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

### OpenRAG Configuration Example

Source: https://context7.com/langflow-ai/openrag/llms.txt

Example environment variables for configuring OpenRAG, including model providers, OpenSearch, and Langflow settings. Ensure to set API keys and passwords appropriately.

```dotenv
OPENAI_API_KEY=sk-...
LLM_PROVIDER=openai
LLM_MODEL=gpt-4o-mini
EMBEDDING_PROVIDER=openai
EMBEDDING_MODEL=text-embedding-3-small

# Alternatively: Anthropic
ANTHROPIC_API_KEY=sk-ant-...
LLM_PROVIDER=anthropic
LLM_MODEL=claude-3-5-sonnet-20241022

# Alternatively: Ollama (local)
OLLAMA_ENDPOINT=http://localhost:11434
LLM_PROVIDER=ollama
EMBEDDING_PROVIDER=ollama

# Alternatively: IBM watsonx.ai
WATSONX_API_KEY=...
WATSONX_ENDPOINT=https://us-south.ml.cloud.ibm.com
WATSONX_PROJECT_ID=...
LLM_PROVIDER=watsonx
EMBEDDING_PROVIDER=watsonx

# OpenSearch (required)
OPENSEARCH_PASSWORD=MyStr0ngP@ssword
OPENSEARCH_INDEX_NAME=documents      # default
OPENSEARCH_HOST=localhost            # default
OPENSEARCH_PORT=9200                 # default

# Langflow
LANGFLOW_SUPERUSER_PASSWORD=...      # recommended
LANGFLOW_PUBLIC_URL=http://localhost:7860

# Cloud connectors (optional)
GOOGLE_OAUTH_CLIENT_ID=...
GOOGLE_OAUTH_CLIENT_SECRET=...
MICROSOFT_GRAPH_OAUTH_CLIENT_ID=...
MICROSOFT_GRAPH_OAUTH_CLIENT_SECRET=...
AWS_ACCESS_KEY_ID=...
AWS_SECRET_ACCESS_KEY=...

# System
FRONTEND_PORT=3000
LOG_LEVEL=INFO
MAX_WORKERS=4
```

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

### Start OpenRAG Services

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/quickstart.mdx

Confirm the prompt to start OpenRAG services. This action pulls and runs necessary container images.

```text
Services are running.
```

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

### Install OpenRAG MCP

Source: https://github.com/langflow-ai/openrag/blob/main/README.md

Install the Model Context Protocol package for OpenRAG using pip. This is required to connect AI assistants to your knowledge base.

```bash
pip install openrag-mcp
```

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

### Quick Start with OpenRAGClient

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/python/README.md

Initialize the OpenRAGClient and perform a simple chat operation. The client automatically discovers API keys and URLs from environment variables.

```python
import asyncio
from openrag_sdk import OpenRAGClient

async def main():
    # Client auto-discovers OPENRAG_API_KEY and OPENRAG_URL from environment
    async with OpenRAGClient() as client:
        # Simple chat
        response = await client.chat.create(message="What is RAG?")
        print(response.response)
        print(f"Chat ID: {response.chat_id}")

asyncio.run(main())
```

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

### Local Development Workflow Setup

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Recommended for active development, this workflow runs infrastructure in Docker while backend and frontend run locally for faster iteration and debugging.

```bash
# Terminal 1: Start infrastructure (OpenSearch, Langflow, Dashboards)
make dev-local-cpu

# Terminal 2: Run backend locally
make backend

# Terminal 3: Run frontend locally
make frontend

# Terminal 4 (optional): Start docling for document processing
make docling
```

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

### Quick Start with OpenRAGClient

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/typescript/README.md

Initialize the client and perform a simple chat operation. The client auto-discovers API key and URL from environment variables.

```typescript
import { OpenRAGClient } from "openrag-sdk";

// Client auto-discovers OPENRAG_API_KEY and OPENRAG_URL from environment
const client = new OpenRAGClient();

// Simple chat
const response = await client.chat.create({ message: "What is RAG?" });
console.log(response.response);
console.log(`Chat ID: ${response.chatId}`);
```

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

### Perform Full Factory Reset

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Execute a complete reset, deleting all data, containers, and volumes for a fresh start.

```bash
make factory-reset
```

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

### Run Continuous Integration Tests

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Perform full CI tests by starting infrastructure, running tests, and tearing down, or build images locally.

```bash
make test-ci
```

```bash
make test-ci-local
```

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

### Clone OpenRAG Repository

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/docker.mdx

Clone the OpenRAG repository to your local machine to begin the setup process.

```bash
git clone https://github.com/langflow-ai/openrag.git
```

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

### Manage Docling Service

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Start and stop the Docling service, which handles document parsing and OCR.

```bash
make docling
```

```bash
make docling-stop
```

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

### Full Docker Stack Management

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Commands to start, stop, and manage the OpenRAG services using Docker. Best for testing the full system.

```bash
make dev          # Start with GPU support
make dev-cpu      # Start with CPU only
make stop         # Stop and remove all containers
```

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

### Operator Build and Deploy Commands

Source: https://github.com/langflow-ai/openrag/blob/main/kubernetes/operator/README.md

A sequence of make commands to download dependencies, generate manifests, compile the operator, install the CRD, and deploy the operator to the cluster.

```bash
make deps          # download controller-gen, kustomize, envtest into ./bin
make manifests     # regenerate CRD + RBAC YAML (run after editing types)
make generate      # regenerate DeepCopy methods (run after editing types)
make build         # compile bin/manager
make install       # install the CRD into the current cluster
make deploy IMG=ghcr.io/langflow-ai/openrag-operator:latest
```

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

### Knowledge Filter Management

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/python/README.md

Examples of creating, searching, getting, updating, and deleting knowledge filters using the OpenRAG SDK.

```APIDOC
## Create Knowledge Filter

### Description
Creates a new knowledge filter with specified parameters.

### Method
`client.knowledge_filters.create(payload)`

### Parameters
#### Request Body
- **name** (string) - Required - The name of the knowledge filter.
- **description** (string) - Optional - A description for the knowledge filter.
- **queryData** (object) - Required - Configuration for the filter query.
  - **query** (string) - Required - The search query string.
  - **filters** (object) - Optional - Additional filtering criteria.
    - **document_types** (array of strings) - Optional - Specifies allowed document types.
  - **limit** (integer) - Optional - The maximum number of results to return.
  - **scoreThreshold** (float) - Optional - The minimum score threshold for results.

### Request Example
```python
result = await client.knowledge_filters.create({
    "name": "Technical Docs",
    "description": "Filter for technical documentation",
    "queryData": {
        "query": "technical",
        "filters": {
            "document_types": ["application/pdf"],
        },
        "limit": 10,
        "scoreThreshold": 0.5,
    },
})
filter_id = result.id
```

## Search Knowledge Filters

### Description
Searches for knowledge filters based on a query string.

### Method
`client.knowledge_filters.search(query)`

### Parameters
#### Query Parameters
- **query** (string) - Required - The search term for filters.

### Request Example
```python
filters = await client.knowledge_filters.search("Technical")
for f in filters:
    print(f"{f.name}: {f.description}")
```

## Get Knowledge Filter

### Description
Retrieves a specific knowledge filter by its ID.

### Method
`client.knowledge_filters.get(filter_id)`

### Parameters
#### Path Parameters
- **filter_id** (string) - Required - The ID of the filter to retrieve.

### Request Example
```python
filter_obj = await client.knowledge_filters.get(filter_id)
```

## Update Knowledge Filter

### Description
Updates an existing knowledge filter.

### Method
`client.knowledge_filters.update(filter_id, payload)`

### Parameters
#### Path Parameters
- **filter_id** (string) - Required - The ID of the filter to update.
#### Request Body
- **description** (string) - Optional - The new description for the filter.

### Request Example
```python
await client.knowledge_filters.update(filter_id, {
    "description": "Updated description",
})
```

## Delete Knowledge Filter

### Description
Deletes a knowledge filter by its ID.

### Method
`client.knowledge_filters.delete(filter_id)`

### Parameters
#### Path Parameters
- **filter_id** (string) - Required - The ID of the filter to delete.

### Request Example
```python
await client.knowledge_filters.delete(filter_id)
```
```

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

### Onboard with Model Providers

Source: https://context7.com/langflow-ai/openrag/llms.txt

Complete initial OpenRAG setup by selecting LLM and embedding models. This configures Langflow, initializes the OpenSearch index, and can optionally ingest sample data.

```bash
# Onboard with OpenAI
curl -X POST http://localhost:3000/settings/onboard \
  -H "Content-Type: application/json" \
  -d '{
    "llm_provider": "openai",
    "llm_model": "gpt-4o-mini",
    "embedding_provider": "openai",
    "embedding_model": "text-embedding-3-small",
    "openai_api_key": "sk-..."
  }'
```

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

### Run MCP from Source

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/mcp/README.md

Navigate to the MCP directory, synchronize dependencies, set environment variables for OpenRAG URL and API key, and then run the MCP server.

```bash
cd sdks/mcp
uv sync
export OPENRAG_URL="http://localhost:3000"
export OPENRAG_API_KEY="orag_your_api_key"
uv run openrag-mcp
```

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

### Install OpenRAG Python Package

Source: https://github.com/langflow-ai/openrag/blob/main/README.md

Install the OpenRAG Python package using pip. Refer to the documentation for detailed installation options.

```bash
pip install openrag-sdk
```

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

### Build Documentation PDF

Source: https://github.com/langflow-ai/openrag/blob/main/docs/README.md

Builds the website and generates a PDF documentation file. This command should be run from the 'openrag/docs' directory.

```bash
npm run build:pdf
```

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

### Access Backend API Documentation

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

When running the backend locally with `make backend`, access the interactive API documentation at this URL.

```bash
make backend
```

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

### Initialize a new uv Python project

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/install-uv.mdx

Create a new Python project managed by uv, specifying the minimum required Python version. Omit the --python flag if your default Python version meets the requirement.

```bash
uv init PROJECT_NAME --python 3.13
```

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

### Build and Serve Docusaurus Site

Source: https://github.com/langflow-ai/openrag/blob/main/docs/VERSIONING_SETUP.md

Run these commands to build the Docusaurus site and serve it locally for testing after configuration changes.

```bash
npm run build
npm run serve
```

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

### Install OpenRAG Plugin Globally

Source: https://github.com/langflow-ai/openrag/blob/main/AGENTS.md

Use these commands to install the OpenRAG plugin globally for Claude Code. This makes the OpenRAG skills available for use.

```bash
/plugin marketplace add langflow-ai/openrag
/plugin install openrag@openrag
```

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

### Access Makefile Help System

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Display help information for various command categories within the Makefile.

```bash
make help
```

```bash
make help_dev
```

```bash
make help_docker
```

```bash
make help_test
```

```bash
make help_local
```

```bash
make help_utils
```

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

### Start OpenRAG GUI Terminal Session

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/get-started/tui.mdx

Start the OpenRAG terminal with the --tui flag for a GUI terminal experience, which includes mouse navigation and button-like controls.

```bash
uv run openrag --tui
```

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

### Initialize OAuth Flow and Handle Callback

Source: https://context7.com/langflow-ai/openrag/llms.txt

Initiate OAuth flows for connecting data sources and handle the callback to complete the authentication process. The /auth/me endpoint retrieves user information.

```bash
# Start an OAuth flow for Google Drive data source
curl -X POST http://localhost:3000/auth/init \
  -H "Content-Type: application/json" \
  -d '{
    "connector_type": "google_drive",
    "purpose": "data_source",
    "name": "My Google Drive",
    "redirect_uri": "http://localhost:3000/auth/callback"
  }'
```

```bash
# Handle OAuth callback
curl -X POST http://localhost:3000/auth/callback \
  -H "Content-Type: application/json" \
  -d '{
    "connection_id": "conn-uuid",
    "authorization_code": "4/0AX4XfWj...",
    "state": "..."
  }'
```

```bash
# Get current user info
curl http://localhost:3000/auth/me
```

```bash
# Logout
curl -X POST http://localhost:3000/auth/logout
```

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

### Remove Old NVIDIA Container Toolkit Packages

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/support/troubleshoot.mdx

Before installing the new NVIDIA Container Toolkit, remove any existing packages to avoid conflicts. This command does not affect your NVIDIA driver installation.

```bash
sudo dnf5 remove -y nvidia-container-toolkit nvidia-container-toolkit-base libnvidia-container*
```

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

### Podman Setup for macOS

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Configure Podman machine with sufficient memory for OpenRAG on macOS. Stop and remove any existing machine before creating a new one with 8GB RAM and 4 CPUs.

```bash
# Stop and remove existing machine (if any)
podman machine stop
podman machine rm

# Create new machine with 8GB RAM and 4 CPUs
podman machine init --memory 8192 --cpus 4
podman machine start
```

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

### Settings - Get

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/python/README.md

Retrieves the current OpenRAG system settings.

```APIDOC
## Settings - Get

### Description
Retrieves the current OpenRAG system settings.

### Method
`client.settings.get()`

### Response Example
```json
{
  "agent": {
    "llm_provider": "openai",
    "llm_model": "gpt-4"
  },
  "knowledge": {
    "embedding_model": "text-embedding-ada-002"
  },
  "chunk_size": 500,
  "chunk_overlap": 100
}
```
```

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

### Chat and Search with Filters

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/python/README.md

Examples of using knowledge filters in chat and search operations.

```APIDOC
## Use Filter in Chat

### Description
Initiates a chat conversation using a specified knowledge filter.

### Method
`client.chat.create(message, filter_id)`

### Parameters
#### Request Body
- **message** (string) - Required - The user's message.
- **filter_id** (string) - Required - The ID of the knowledge filter to apply.

### Request Example
```python
response = await client.chat.create(
    message="Explain the API",
    filter_id=filter_id,
)
```

## Use Filter in Search

### Description
Performs a search query using a specified knowledge filter.

### Method
`client.search.query(query, filter_id)`

### Parameters
#### Request Body
- **query** (string) - Required - The search query.
- **filter_id** (string) - Required - The ID of the knowledge filter to apply.

### Request Example
```python
results = await client.search.query("API endpoints", filter_id=filter_id)
```
```

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

### Create Subsequent Docs Version

Source: https://github.com/langflow-ai/openrag/blob/main/docs/VERSIONING_SETUP.md

Repeat the CLI command to create additional versions (e.g., 2.0.0) from the current documentation.

```bash
# Create version 2.0.0 from current docs
npm run docusaurus docs:version 2.0.0
```

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

### Chat - Get Conversation

Source: https://github.com/langflow-ai/openrag/blob/main/sdks/python/README.md

Retrieves a specific chat conversation, including its messages, by its chat_id.

```APIDOC
## Chat - Get Conversation

### Description
Retrieves a specific chat conversation, including its messages, by its chat_id.

### Method
`client.chat.get(chat_id: str)`

### Parameters
- **chat_id** (str) - Required - The ID of the conversation to retrieve.

### Response Example
```json
{
  "chat_id": "conversation_id_123",
  "title": "Conversation Title",
  "messages": [
    {
      "role": "user",
      "content": "User message content"
    },
    {
      "role": "assistant",
      "content": "Assistant response content"
    }
  ]
}
```
```

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

### Navigate to Docs Directory

Source: https://github.com/langflow-ai/openrag/blob/main/docs/docs/support/contribute.mdx

Change to the `/docs` directory in your local OpenRAG repository. If using a development container, run documentation builds from your host terminal.

```bash
cd openrag/docs
```

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

### Manage Settings with OpenRAG Python SDK

Source: https://github.com/langflow-ai/openrag/blob/main/plugins/openrag/skills/sdk/SKILL.md

Demonstrates retrieving current system settings and updating them, including LLM and embedding provider/model configurations, using the Python SDK. Ensure you have the necessary permissions to update settings.

```python
async def manage_settings():
    async with OpenRAGClient() as client:
        # Get settings
        settings = await client.settings.get()
        print(f"LLM Provider: {settings.agent.llm_provider}")
        print(f"LLM Model: {settings.agent.llm_model}")
        print(f"Embedding Model: {settings.knowledge.embedding_model}")
        
        # Update settings
        await client.settings.update({
            "llm_provider": "openai",
            "llm_model": "gpt-4o",
            "embedding_provider": "openai",
            "embedding_model": "text-embedding-3-small"
        })

asyncio.run(manage_settings())
```

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

### Deploy Website Using SSH

Source: https://github.com/langflow-ai/openrag/blob/main/docs/README.md

Deploys the website using SSH. Ensure SSH is configured for your deployment environment.

```bash
USE_SSH=true npm run deploy
```

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

### Get current user info

Source: https://context7.com/langflow-ai/openrag/llms.txt

Retrieves information about the currently authenticated user and the OpenRAG version.

```APIDOC
## GET /auth/me

### Description
Returns details about the current user, including their ID, name, email, and the OpenRAG version.

### Method
GET

### Endpoint
/auth/me

### Response Example (Success)
```json
{
  "user_id": "user123",
  "name": "Jane Doe",
  "email": "jane@example.com",
  "version": "0.3.2"
}
```
```

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

### Verify OpenRAG Container Status

Source: https://github.com/langflow-ai/openrag/blob/main/plugins/openrag/skills/install/SKILL.md

After installation, check if the OpenRAG container is running using this Docker command.

```bash
docker ps --filter "name=openrag"
```

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

### Run Backend and Integration Tests

Source: https://github.com/langflow-ai/openrag/blob/main/CONTRIBUTING.md

Execute backend tests, integration tests, or SDK tests. Linting checks are also available.

```bash
make test
```

```bash
make test-integration
```

```bash
make test-sdk
```

```bash
make lint
```

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

### TypeScript/JavaScript SDK Usage

Source: https://context7.com/langflow-ai/openrag/llms.txt

Examples of using the OpenRAG SDK in TypeScript/JavaScript for chat, search, and document ingestion.

```APIDOC
## TypeScript/JavaScript SDK

### Description
This section demonstrates how to use the `openrag-sdk` in TypeScript/JavaScript for various OpenRAG functionalities.

### Methods
- `client.chat.create({ message: string, chatId?: string })`: For sending messages and managing chat conversations.
- `client.search.query({ query: string, limit?: number, scoreThreshold?: number })`: To perform semantic searches.
- `client.documents.ingest({ file: ReadableStream, filename: string })`: To upload and ingest documents.
- `client.tasks.get(taskId: string)`: To check the status of ingestion tasks.

### Example Usage
```typescript
import { OpenRAGClient } from "openrag-sdk";

const client = new OpenRAGClient({
  baseUrl: "http://localhost:3000",
  apiKey: process.env.OPENRAG_API_KEY,
});

// Chat
const response = await client.chat.create({ message: "Summarize the onboarding docs" });
console.log(response.response);

// Multi-turn
const followup = await client.chat.create({
  message: "Which step is most important?",
  chatId: response.chatId,
});
console.log(followup.response);

// Semantic search
const searchResults = await client.search.query({
  query: "data retention policy",
  limit: 5,
  scoreThreshold: 0.5,
});
for (const r of searchResults.results) {
  console.log(`[${r.score}] ${r.filename}: ${r.text.slice(0, 80)}`);
}

// Ingest a document (Node.js)
import fs from "fs";
const fileStream = fs.createReadStream("/path/to/document.pdf");
const task = await client.documents.ingest({ file: fileStream, filename: "document.pdf" });

// Poll task
let taskStatus = await client.tasks.get(task.taskId);
while (!["completed", "failed"].includes(taskStatus.status)) {
  await new Promise((r) => setTimeout(r, 2000));
  taskStatus = await client.tasks.get(task.taskId);
}
console.log("Final status:", taskStatus.status);
```
```