A Retrieval Augmented Generation (RAG) agent built with Google's Agent Development Kit (ADK) and Vertex AI. Query your documents using natural language and get AI-powered answers.

The Vertex AI RAG Agent allows you to:

• Query document corpora with natural language questions
• List available document corpora
• Create new document corpora
• Add new documents to existing corpora
• Get detailed information about specific corpora
• Delete corpora when they're no longer needed

flowchart TB User["👤 User"] Agent["🤖 RAG Agent"] Tools["🔧 Python Tools"] VertexRAG["☁️ Vertex AI RAG API"] User --> Agent Agent --> Tools Agent --> VertexRAG 

• ✅ Google Cloud Account with billing enabled
• ✅ Google Cloud Project with Vertex AI API enabled
• ✅ Python 3.9+ and gcloud CLI installed
• ✅ IAM permissions for Vertex AI resources

1. Authenticate with Google Cloud

gcloud auth application-default login gcloud services enable aiplatform.googleapis.com

2. Install Dependencies

pip install -r requirements.txt

3. Configure Your Project

Create a .env file with your project details:

cp .env.example .env

Edit .env:

GOOGLE_CLOUD_PROJECT=your-project-id GOOGLE_CLOUD_LOCATION=us-central1

4. Run the Agent

adk web

Done! 🎉 Your RAG agent is ready.

Start the agent interface:

# Web interface (recommended) adk web # Command line interface adk run

1. Create and Query a Knowledge Base

Create a corpus called 'product_docs' 

Add these documents to product_docs: - https://drive.google.com/file/d/abc123/view - gs://my-bucket/docs/manual.pdf 

What are the main features of the product? 

2. Search Across Documents

What does the documentation say about installation? 

3. Manage Your Collections

List all available corpora 

Show me details about the product_docs corpus 

Delete the old_corpus corpus 

• 📚 Knowledge Base - Company documentation search
• 🔬 Research Assistant - Query academic papers
• 💬 Customer Support - Answer questions from support docs
• 📋 Legal Analysis - Search contracts and legal documents

Customize RAG settings in rag_agent/config.py:

Uses Google Cloud Application Default Credentials (ADC). Run gcloud auth application-default login before starting the agent.

Required IAM Permissions:

• aiplatform.ragCorpora.* (create, delete, get, list, update)
• storage.objects.get (for GCS access)
• drive.files.read (for Google Drive access)

• Vertex AI RAG Docs - Official Vertex AI documentation
• Google ADK Docs - Agent Development Kit documentation