feat(aisearch): aisearch-action guide

Author: danielebriggi

sqlite-cloud/aisearch-docs.mdx

@@ -0,0 +1,67 @@
+---
+title: AI Search for your documents and files
+description: SQLite AI Search for your documents and files
+category: getting-started
+status: publish
+slug: aisearch-documents
+---
+
+# AI Search for Your Documents and Files
+
+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](https://github.com/sqliteai/sqlite-aisearch-action), you can integrate document processing into your GitHub workflow and set up a chat bot on your site in just a few steps.
+
+The semantic search is powered by [SQLite RAG](https://github.com/sqliteai/sqlite-rag).
+
+## Set Up Your GitHub Workflow
+
+1. **Get Your Connection String**: Ensure you have a project on the [SQLite Cloud dashboard](https://dashboard.sqlitecloud.io). If not, sign up for [SQLite AI](https://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:
+
+```yaml
+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
+```
+
+## 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](@docs-website-assets/aisearch-docs/edgefn_aisearch.png)
+4. Create a new `Javascript Function` and copy the code from [aisearch-docs.js](search_edge_function_template/aisearch-docs.js) into the editor
+5. Deploy and test
+
+### How to Perform a Search
+
+1. Go to **Details** in the Edge Function panel and copy the **Function URL**
+![AISearch Edge Function Details](@docs-website-assets/aisearch-docs/edgefn_aisearch_details.png)
+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
+    ```