// customize
const inkeepSettings = {
    isOpen: true,
    baseSettings: {
        apiKey: '3adc15f45617826bb73c0a4cb9fb3a06547b6ab7f90c819e',
        integrationId: 'clv1yuc8f0002vzhxbzwy26yx',
        organizationId: 'clu6yudh9000sajgn7nwxfve8',
        primaryBrandColor: '#26D6FF', // your brand color, widget color scheme is derived from this
        organizationDisplayName: 'R2R',
        theme: {
          colorMode: {
            forcedColorMode: 'dark', // to sync dark mode with the widget
          },
        },
    },
    aiChatSettings: {
      // ...optional settings
      botAvatarSrcUrl:
        "https://www.sciphi.ai/screenshots/logo222_cut.png",
      quickQuestions: [
        "How do I get started?",
        "How does R2R implement hybrid search?",
        "How do I use the R2R API?",
        // "How do I use ",
        // "Example question 3?",
      ],
    },
    modalSettings: {
      isShortcutKeyEnabled: false, // disable default cmd+k behavior
      // ...optional settings
    },
  };

  // The Mintlify search triggers, which we'll reuse to trigger the Inkeep modal
  const searchButtonContainerIds = [
    "search-bar-entry",
    "search-bar-entry-mobile",
  ];

  // Clone and replace, needed to remove existing event listeners
  const clonedSearchButtonContainers = searchButtonContainerIds.map((id) => {
    const originalElement = document.getElementById(id);
    const clonedElement = originalElement.cloneNode(true);
    originalElement.parentNode.replaceChild(clonedElement, originalElement);

    return clonedElement;
  });

  // Load the Inkeep component library
  const inkeepScript = document.createElement("script");
  inkeepScript.type = "module";
  inkeepScript.src =
    "https://unpkg.com/@inkeep/widgets-embed@latest/dist/embed.js";
  document.body.appendChild(inkeepScript);

  // Once the Inkeep library is loaded, instantiate the UI components
  inkeepScript.addEventListener("load", function () {
    // Customization settings

    // for syncing with dark mode
    const colorModeSettings = {
      observedElement: document.documentElement,
      isDarkModeCallback: (el) => {
        return el.classList.contains("dark");
      },
      colorModeAttribute: "class",
    };

    // Instantiate the "Ask AI" pill chat button
    Inkeep().embed({
      componentType: "ChatButton",
      colorModeSync: colorModeSettings,
      properties: inkeepSettings,
    });

    // Instantiate the search bar modal
    const inkeepSearchModal = Inkeep({
      ...inkeepSettings.baseSettings,
    }).embed({
      componentType: "CustomTrigger",
      colorModeSync: colorModeSettings,
      properties: {
        ...inkeepSettings,
        isOpen: false,
        onClose: () => {
          inkeepSearchModal.render({
            isOpen: false,
          });
        },
      },
    });

    // When the Mintlify search bar elements are clicked, open the Inkeep search modal
    clonedSearchButtonContainers.forEach((trigger) => {
      trigger.addEventListener("click", function () {
        inkeepSearchModal.render({
          isOpen: true,
        });
      });
    });

    // Open the Inkeep Modal with cmd+k
    window.addEventListener(
      "keydown",
      (event) => {
        if (
          (event.metaKey || event.ctrlKey) &&
          (event.key === "k" || event.key === "K")
        ) {
          event.stopPropagation();
          inkeepSearchModal.render({ isOpen: true });
          return false;
        }
      },
      true
    );
  });


Creating a graph on your documents. This endpoint takes input a list of document ids and KGCreationSettings. If document IDs are not provided, the graph will be created on all documents in the system.
This step extracts the relevant entities and relationships from the documents and creates a graph based on the extracted information.
In order to do GraphRAG, you will need to run the enrich_graph endpoint.

Create Graph

Build, scale, and manage user-facing Retrieval-Augmented Generation applications.

Introduction

System

What's New

Quickstart

Learn how to run a Retrieval-Augmented Generation system locally using R2R

Local LLMs

Powerful document ingestion, search, and RAG capabilities at your fingertips

R2R API Reference

Ingest files into the system.

This endpoint supports multipart/form-data requests, enabling you to ingest files and their associated metadatas into R2R.

A valid user authentication token is required to access this endpoint, as regular users can only ingest files for their own access. More expansive collection permissioning is under development.

Ingest Files

Ingest text chunks into the system.

This endpoint supports multipart/form-data requests, enabling you to ingest pre-parsed text chunks into R2R.

A valid user authentication token is required to access this endpoint, as regular users can only ingest chunks for their own access. More expansive collection permissioning is under development.

Ingest Chunks

Update existing files in the system.

This endpoint supports multipart/form-data requests, enabling you to update files and their associated metadatas into R2R.

A valid user authentication token is required to access this endpoint, as regular users can only update their own files. More expansive collection permissioning is under development.

Update Files

This endpoint enriches the graph with additional information. It creates communities of nodes based on their similarity and adds embeddings to the graph. This step is necessary for GraphRAG to work.

Enrich Graph

Perform a search query on the vector database and knowledge graph.

This endpoint allows for complex filtering of search results using PostgreSQL-based queries.
Filters can be applied to various fields such as document_id, and internal metadata values.


Allowed operators include `eq`, `neq`, `gt`, `gte`, `lt`, `lte`, `like`, `ilike`, `in`, and `nin`.

Search

Execute a RAG (Retrieval-Augmented Generation) query.

This endpoint combines search results with language model generation.
It supports the same filtering capabilities as the search endpoint,
allowing for precise control over the retrieved context.

The generation process can be customized using the rag_generation_config parameter.

Implement an agent-based interaction for complex query processing.

This endpoint supports multi-turn conversations and can handle complex queries
by breaking them down into sub-tasks. It uses the same filtering capabilities
as the search and RAG endpoints for retrieving relevant information.

The agent's behavior can be customized using the rag_generation_config and
task_prompt_override parameters.

Agent

Document Chunks

Delete

Documents Overview

Register a new user with the given email and password.

Register

Verify a user's email address.

This endpoint is used to confirm a user's email address using the verification code
sent to their email after registration.

Verify Email

Authenticate a user and provide access tokens.

This endpoint authenticates a user using their email (username) and password,
and returns access and refresh tokens upon successful authentication.

Login

Get the current user's profile information.

This endpoint returns the profile information of the currently authenticated user.

Get User

Update the current user's profile information.

This endpoint allows the authenticated user to update their profile information.

Update User

Delete a user account.

This endpoint allows users to delete their own account or, for superusers,
to delete any user account.

Delete User

Refresh the access token using a refresh token.

This endpoint allows users to obtain a new access token using their refresh token.

Refresh Access Token

Log out the current user.

This endpoint invalidates the user's current access token, effectively logging them out.

Logout

Change the authenticated user's password.

This endpoint allows users to change their password by providing their current password
and a new password.

Change Password

Request a password reset for a user.

This endpoint initiates the password reset process by sending a reset link
to the specified email address.

Password Reset

Create Collection

Get Collection

Update Collection

Delete Collection

List Collections

Add User to Collection

Remove User from Collection

Get Users in Collection

Get Collections for User

Collections Overview

Assign Document to Collection

Remove Document from Collection

Get Document Collections

Update Prompt

Add Prompt

Get Prompt

Get All Prompts

Delete Prompt

Logs

Users Overview

App Settings

Analytics

Score Completion

Health

A detailed step-by-step cookbook of the core features provided by R2R.

Walkthrough

Learn how to implement and use hybrid search with R2R

Hybrid Search

Learn how to build and use GraphRAG with R2R

GraphRAG

Learn how to build and use advanced RAG techniques with R2R

Advanced RAG

Learn how to implement and use agents with R2R

Agents

Learn how orchestration is handled inside R2R

Orchestration

Learn how to build webapps powered by RAG using R2R

Web Development

A comprehensive guide to user authentication and management features in R2R

User Auth

A comprehensive guide to creating collections in R2R

Collections

Learn how to use analytics and and logging with R2R

Analytics & Observability

Learn how to set up and use the R2R Application for managing your instance.

Application

Learn how to deploy R2R into remote cloud environments

Remote Cloud

Learn how to deploy an R2R project in the cloud

Deploy

Interact with deployed projects from the cloud

Projects

Developer Quickstart

Security

Community

Documentation

API Reference

Cookbooks

SciPhi Cloud

R2R GitHub

Support

Learn how to configure your R2R deployment

Postgres

Learn how to configure LLMs in your R2R deployment

LLMs

Learn how to configure RAG in your R2R deployment

Learn how to configure and manage prompts in your R2R deployment

Prompts

Overview

Learn how to configure document chunking in your R2R deployment

Parsing & Chunking

Embedding

Configure your R2R knowledge graph provider.

Configuration for Restructuring data after ingestion using Knowledge Graphs

KG Enrichment

Vector Search

Knowledge Graph

Azure

Learn how to deploy R2R into Google Cloud

Core Logic

R2RConfig

R2RBuilder

Learn how to configure and use the authentication provider in R2R

Crypto & Auth

Configure and use multiple Language Model providers in R2R

Language Models (LLMs)

Learn how to configure and use the database provider in R2R

Database

Learn how to configure and use the knowledge graph provider in R2R

Learn about R2R telemetry and how to manage it

API Documentation

Document Ingestion

Knowledge Graph

Retrieval

User Documents

User Auth

Collections

Admin

Analytics and Observability

Status

Create Graph