SQLite Cloud

Build AI Search for Your Documentation

This guide shows you how to set up a ready-to-use AI semantic search for your documents and files. Using the sqlite-aisearch-action, you can integrate document processing into your GitHub workflow and set up a chatbot on your site in just a few steps.

The semantic search is powered by SQLite RAG.

Step 1: Set Up Your GitHub Workflow

  1. Get Your Connection String: Ensure you have a project on the SQLite Cloud dashboard. If not, sign up for SQLite AI to create one for free.

  2. Set GitHub Secret: Add your connection string as SQLITECLOUD_CONNECTION_STRING in your repository secrets.

  3. Add to Workflow: Create or update your GitHub workflow:

name: AI Search Index

on:
  push:
    branches: [main]
  workflow_dispatch:

jobs:
  build-search:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Build AI Search Database
        uses: sqliteai/sqlite-aisearch-action@v1
        with:
          connection_string: ${{ secrets.SQLITECLOUD_CONNECTION_STRING }}
          base_url: https://docs.yoursite.com
          database_name: aidocs_search.db
          source_files: ./path/to/documents

Step 2: Create the Search Edge Function

To enable search functionality on your indexed database, create an Edge Function using the provided template:

  1. Access your dashboard at https://dashboard.sqlitecloud.io
  2. Navigate to the same project where your database was uploaded
  3. Go to the Edge Functions section AISearch Edge Function
  4. Create a new Javascript Function and copy the code from aisearch-docs.js into the editor
  5. Deploy and test

  1. Go to Details in the Edge Function panel and copy the Function URL AISearch Edge Function Details

  2. Execute a GET request with a URL-encoded query as the query parameter

    Example:

    GET https://myproject.cloud/v2/functions/aisearch-docs?query=what%27s+Offsync%3F

Step 3: Integrate the Docs Chatbot in Your Website

Once you have your search edge function deployed, you can easily add an interactive AI chatbot to your website. The chatbot provides a user-friendly interface for your documentation search, powered by the indexed content.

Docs Chatbot

React Integration

Install the chatbot package:

npm install @sqliteai/docs-chatbot

Then add it to your React application:

import { DocsChatbot } from "@sqliteai/docs-chatbot";
import "@sqliteai/docs-chatbot/style.css";

function App() {
  return (
    <DocsChatbot
      searchUrl="https://yourproject.sqlite.cloud/v2/functions/aisearch-docs"
      apiKey="your-api-key"
      title="Help Center"
    />
  );
}

  • Replace the searchUrl with your Function URL from Step 2
  • For the apiKey, you need to use an API key with read permissions on your AI docs database. Learn how to create and manage API keys in the API Key documentation.

Vanilla JavaScript

For non-React applications, use the web component:

<!DOCTYPE html>
<html lang="en">
  <head>
    <meta charset="UTF-8" />
    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
  </head>
  <body>
    <script src="https://unpkg.com/@sqliteai/docs-chatbot/dist/umd/docs-chatbot.min.js"></script>

    <docs-chatbot
      search-url="https://yourproject.sqlite.cloud/v2/functions/aisearch-docs"
      api-key="your-api-key"
      title="Help Center"
    >
    </docs-chatbot>
  </body>
</html>

By default, the chatbot displays as a floating button in the bottom-right corner. For advanced configuration options including custom triggers, theming, and API reference, see the full docs chatbot documentation.