AgenticRAG
AgenticRAG is a modular library for building customizable Retrieval-Augmented Generation systems. It provides a complete stack for ingesting, storing, retrieving, and acting on diverse data types through a unified interface.
Key Features
- Modular Architecture: Easily customize any component while maintaining system compatibility
- Multi-format Data Support: Handle text, tables, and databases through specialized stores
- Flexible Data Ingestion: Import data from PDFs, CSVs, web pages, and external databases
- Intelligent Retrieval: Use semantic search, SQL queries, or table operations as needed
- Task-oriented Design: Perform question answering, chart generation, or custom operations
- Easy Extensibility: Create custom components with minimal boilerplate
Installation
For Users
Install the stable release from PyPI:
pip install agenticrag
For Contributors
Clone the repository and install development dependencies:
git clone https://github.com/yourusername/agenticrag.git
cd agenticrag
pip install -r requirements.txt
Quick Start
from agenticrag import RAGAgent
# Initialize agent with default components
agent = RAGAgent(persistent_dir="./agenticrag_data")
# Load some data
agent.load_pdf("path/to/document.pdf", name="company_handbook")
agent.load_csv("path/to/data.csv", name="sales_data")
# Ask questions using the data
response = agent.invoke("What were our top selling products last quarter?")
print(response)
Architecture Overview
The AgenticRAG system follows a modular, layered architecture where each component has specific responsibilities:
- Data Storage Layer: Stores maintain structured representations of various data types
- Data Ingestion Layer: Loaders and Connectors handle importing data into stores
- Data Retrieval Layer: Retrievers access relevant information based on queries
- Task Execution Layer: Tasks perform operations using retrieved context
- Orchestration Layer: RAGAgent coordinates all components to fulfill user requests
Documentation
For detailed documentation, check out:
- Full Documentation: Complete guide to architecture and components
- Examples: Jupyter notebooks showing various use cases
Project Structure
├── agenticrag/ # Main package
│ ├── connectors/ # Database and API connectors
│ ├── loaders/ # Data ingestion components
│ ├── retrievers/ # Context retrieval components
│ ├── stores/ # Data storage components
│ ├── tasks/ # Task execution components
│ ├── types/ # Data models and type definitions
│ ├── utils/ # Utility functions and helpers
│ └── rag_agent.py # Main agent implementation
├── docs/ # Documentation
├── examples/ # Example notebooks
├── tests/ # Unit and integration tests
├── ui/ # Streamlit demo app
├── LICENSE # MIT License
├── pyproject.toml # Package configuration
├── README.md # This file
└── requirements.txt # Development dependencies
Example Use Cases
Question Answering from Documents
from agenticrag import RAGAgent
agent = RAGAgent()
agent.load_pdf("research_paper.pdf", name="research")
agent.invoke("Summarize the key findings of the research paper")
Data Analysis with Visualizations
from agenticrag import RAGAgent
from agenticrag.retrievers import TableRetriever
from agenticrag.tasks import ChartGenerationTask
agent = RAGAgent(
retrievers=[TableRetriever()],
tasks=[ChartGenerationTask()]
)
agent.load_csv("Iris.csv")
agent.invoke("Create a scatter plot showing petel length and width of Iris")
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add some amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
License
This project is licensed under the MIT License - see the LICENSE file for details.
Metadata
Release files for agenticrag 0.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| agenticrag-0.2.tar.gz | 4.4 kB | Without provenance | Details for agenticrag-0.2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agenticrag-0.2-py3-none-any.whl | Python 3 | none | any | Details for agenticrag-0.2-py3-none-any.whl |
Total release size: 9.2 kB
Release files / agenticrag-0.2.tar.gz
| Download URL | agenticrag-0.2.tar.gz |
|---|---|
| Size | 4.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
e0faa1de0c4fb55bf858736286a3437478d8a395408310706eb5d1aafe717d0e
|
|
BLAKE2b-256 checksum How to use checksums |
1eef53ee380a8be9deac448ce3a95758d213fe216267aba971b3c323896db047
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.16
|
Release files / agenticrag-0.2-py3-none-any.whl
| Download URL | agenticrag-0.2-py3-none-any.whl |
|---|---|
| Size | 4.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
88e09651aadb6940412ed16f84f0f5cf7d7300316c242ee9de3dad5e3484c1b0
|
|
BLAKE2b-256 checksum How to use checksums |
6db3478ce0fd0a44dee68679c83445c74a5d551d8cb213f84eb43e33b73ad5b2
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.16
|
Security for agenticrag 0.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.