Build an agentic RAG app from scratch with Claude Code. Learn hybrid search, reranking, text-to-SQL, and subagents. Use React, FastAPI, and Supabase. Ideal for operations teams automating data workflows.
git clone https://github.com/theaiautomators/claude-code-agentic-rag-masterclass.gitThis masterclass teaches you to build a complete agentic RAG application by collaborating with Claude Code rather than writing code yourself. You'll implement a full pipeline including document ingestion, hybrid search with reranking, text-to-SQL capabilities, and subagent patterns. The course covers both frontend (React, TypeScript, shadcn/ui) and backend (FastAPI, Python) layers, with Supabase handling storage, authentication, and vector embeddings. It's designed for technically-minded professionals without coding experience who need to automate data workflows and build intelligent document retrieval systems.
Clone the repository, install Claude Code in your IDE (Cursor or VS Code), then run `claude` in the terminal and use the `/onboard` command to begin. Follow the 8 modules sequentially using the PRD.md, CLAUDE.md, and PROGRESS.md documentation provided. Claude Code will generate the application as you guide it through each feature.
Build internal knowledge bases with intelligent search and chat interfaces
Automate data workflows using text-to-SQL and web search agents
Create document processing pipelines supporting PDFs, Word docs, HTML, and Markdown
Implement multi-document analysis with isolated subagents for specialized tasks
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/theaiautomators/claude-code-agentic-rag-masterclassCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
I want to build an agentic RAG app using Claude Code. My app will help [COMPANY] in the [INDUSTRY] sector automate [DATA] workflows. Guide me through setting up hybrid search, reranking, text-to-SQL, and subagents using React, FastAPI, and Supabase. Provide step-by-step instructions with code snippets.
# Building an Agentic RAG App for [COMPANY] in the [INDUSTRY] Sector
## Step 1: Setting Up the Project
1. **Initialize the Project**:
```bash
mkdir agentic-rag-app
cd agentic-rag-app
npm init -y
npm install react fastapi supabase
```
2. **Set Up Supabase**:
- Create a new project in Supabase.
- Note your project URL and API key.
## Step 2: Implementing Hybrid Search
1. **Integrate Claude Code**:
```python
from claude_code import ClaudeCode
claude = ClaudeCode(api_key='your_api_key')
```
2. **Hybrid Search Function**:
```python
def hybrid_search(query, documents):
# Implement hybrid search logic here
return reranked_results
```
## Step 3: Adding Reranking and Text-to-SQL
1. **Reranking Logic**:
```python
def rerank_results(results):
# Implement reranking logic here
return reranked_results
```
2. **Text-to-SQL Conversion**:
```python
def text_to_sql(query):
# Implement text-to-SQL logic here
return sql_query
```
## Step 4: Creating Subagents
1. **Define Subagents**:
```python
class Subagent:
def __init__(self, task):
self.task = task
def execute(self):
# Implement subagent logic here
return result
```
2. **Integrate Subagents**:
```python
subagent = Subagent(task='data_analysis')
result = subagent.execute()
```AI assistant built for thoughtful, nuanced conversation
PostgreSQL backend-as-a-service with real-time APIs
AI-assisted web application security testing
AI sales agent for lead generation and follow-up
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan