IAToolkit
Project description
IAToolkit
The Open-Source Framework for Building AI Assistants on Your Private Data
✨ Why IAToolkit?
IAToolkit is more than a framework — it's a complete foundation for building enterprise-grade AI assistants capable of understanding your proprietary data, automating workflows, and integrating seamlessly with your existing systems.
Whether you’re building a production chatbot for your company or learning how real AI applications are architected, IAToolkit gives you everything you need to build secure, production-grade assistants — from data access, multi-tenant architecture, and tool execution to prompt orchestration, semantic search, and full UI integration — all powered by a clean, extensible Python API.
🔌 Connect to Anything
Build AI assistants that truly understand your business.
- Connect to SQL databases (PostgreSQL, MySQL, SQLite)
- Query structured data using natural language
- Perform semantic search on PDFs, DOCX, TXT, XLSX
- Works as a full RAG engine out-of-the-box
🏢 Multi-Tenant by Design
Ideal for SaaS, agencies, consultancies, or large enterprises.
- Isolated Company Modules
- Each company defines its own:
- data sources
- prompts
- custom tools
- branding + UI theme
- Clean separation for multi-client deployments
- Perfect for scaling from 1 to 100 customers
- Configure everything using simple YAML
🧩 Extensible & Provider-Agnostic
Customize every layer and keep full control.
- Add custom tools the LLM can call (SQL, API, Python functions)
- Swap between OpenAI (GPT), Google Gemini, or future models
- Built on dependency injection for maximum modularity
- Extend prompts, dispatchers, services, and LLM clients effortlessly
🛡️ Built for Production
Designed for real-world systems — reliable, maintainable, and scalable.
- Integrated authentication and session handling
- Secure secret management via environment variables
- Detailed logging of prompts, tool calls, and token usage
- Structured architecture with 90%+ automated test coverage
- Clean separation of UI, business logic, and LLM orchestration
- Runs anywhere: local machine, docker, cloud, serverless
🎓 Learn How Real AI Applications Are Built
IAToolkit is also an excellent learning resource:
- Understand how to orchestrate multiple LLM calls
- Learn how tools, SQL, vector search and prompts interact
- Explore a complete architecture for real AI products
- Use the Sample Company demo to practice building AI assistants
- Perfect for developers, students, and teams exploring enterprise AI
🚀 Getting Started in 3 Minutes
Get your first AI assistant running locally by following our "Hello World" example.
Our detailed guide will walk you through setting up your virtual environment, configuring your .env file,
and launching the application.
📚 Documentation
Our documentation is designed for users of all levels, from initial setup to advanced development.
| Guide | Description |
|---|---|
| 🚀 Quickstart Guide | The fastest way to install, configure, and run IAToolkit for the first time. |
| ⚙️ Companies & Components | A deep dive into the company.yaml file, the core of all configuration. |
| 💻 Programming Guide | Understand the internal architecture, services, and design patterns to extend the framework. |
| ☁️ Deployment Guide | Learn how to deploy your IAToolkit application to a production environment. |
| 🗃️ Database Guide | An overview of the core database schema used by the IAToolkit framework itself. |
🤝 Contributing
We welcome contributions of all kinds! Whether it's a new feature, a bug fix, or an improvement to the documentation, your help is appreciated. Please read our Contributing Guide to get started.
📄 License
IAToolkit is open-source software licensed under the MIT License.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file iatoolkit-0.77.0.tar.gz.
File metadata
- Download URL: iatoolkit-0.77.0.tar.gz
- Upload date:
- Size: 258.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
ff088199e888a4061504f3e81709b95a0439b7bbde1d1971391741aa2296024f
|
|
| MD5 |
3b01b0b1cab5c3b3d3e2f979d3a5ffeb
|
|
| BLAKE2b-256 |
a929c07d48e432e9d69a9e67e37000e255d5251a588efe517185ae58676567ef
|
Provenance
The following attestation bundles were made for iatoolkit-0.77.0.tar.gz:
Publisher:
publish.yml on flibedinsky/iatoolkit
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
iatoolkit-0.77.0.tar.gz -
Subject digest:
ff088199e888a4061504f3e81709b95a0439b7bbde1d1971391741aa2296024f - Sigstore transparency entry: 720375684
- Sigstore integration time:
-
Permalink:
flibedinsky/iatoolkit@b773cd66d70705e41669f81298718aa039deb977 -
Branch / Tag:
refs/tags/v0.77.0 - Owner: https://github.com/flibedinsky
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@b773cd66d70705e41669f81298718aa039deb977 -
Trigger Event:
push
-
Statement type:
File details
Details for the file iatoolkit-0.77.0-py3-none-any.whl.
File metadata
- Download URL: iatoolkit-0.77.0-py3-none-any.whl
- Upload date:
- Size: 307.7 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
7802bda76b47038c40a6814d0df373a974a816a9560f38015a41e7487ff7697c
|
|
| MD5 |
a0a714d8e4f67d5432d496ddd04d7a86
|
|
| BLAKE2b-256 |
06ff7a3b67dbb62fa46aa1bd8f3519fbb98d9f00fd1ba935ed58e21a79a0297b
|
Provenance
The following attestation bundles were made for iatoolkit-0.77.0-py3-none-any.whl:
Publisher:
publish.yml on flibedinsky/iatoolkit
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
iatoolkit-0.77.0-py3-none-any.whl -
Subject digest:
7802bda76b47038c40a6814d0df373a974a816a9560f38015a41e7487ff7697c - Sigstore transparency entry: 720375737
- Sigstore integration time:
-
Permalink:
flibedinsky/iatoolkit@b773cd66d70705e41669f81298718aa039deb977 -
Branch / Tag:
refs/tags/v0.77.0 - Owner: https://github.com/flibedinsky
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@b773cd66d70705e41669f81298718aa039deb977 -
Trigger Event:
push
-
Statement type: