title: HashiruAI emoji: 🍆 colorFrom: green colorTo: yellow sdk: gradio sdkversion: 5.31.0 pythonversion: 3.11.9 app_file: start.py

pinned: false

HASHIRU: Hierarchical Agent System for Hybrid Intelligent Resource Utilization

Project Overview

This project provides a framework for creating and managing AI agents and tools. It includes features for managing resource and expense budgets, loading tools and agents, and interacting with various language models.

The architecture consists of a hierarchical system where a CEO agent manages a set of employee agents and tools. Both agents and tools can be autonomously created, invoked, and deleted by the CEO. Agents are associated with reclaimable costs (if local) and non-reclaimable costs (if cloud-based).

The project is designed to be modular and extensible, allowing users to integrate their own tools and agents. It supports multiple language model integrations, including Ollama, Gemini, Groq, and Lambda Labs.

NOTE: Benchmarking efforts of the HASHIRU architecture can be found in HASHIRUBench.

Directory Structure

• src/: Contains the source code for the project.
  • tools/: Contains the code for the tools that can be used by the agents.
    • default_tools/: Contains the default tools provided with the project.
    • user_tools/: Contains the tools created by the user.
  • config/: Contains configuration files for the project.
  • utils/: Contains utility functions and classes used throughout the project.
  • models/: Contains the configurations and system prompts for the agents. Includes models.json which stores agent definitions.
  • manager/: Contains the core logic for managing agents, tools, and budgets.
    • agent_manager.py: Manages the creation, deletion, and invocation of AI agents. Supports different agent types like Ollama, Gemini, and Groq.
    • budget_manager.py: Manages the resource and expense budgets for the project.
    • tool_manager.py: Manages the loading, running, and deletion of tools.
    • llm_models.py: Defines abstract base classes for different language model integrations.
  • data/: Contains data files, such as memory and secret words.

Key Components

• Agent Management: The AgentManager class in src/manager/agent_manager.py is responsible for creating, managing, and invoking AI agents. It supports different agent types, including local (Ollama) and cloud-based (Gemini, Groq) models.
• Tool Management: The ToolManager class in src/manager/tool_manager.py handles the loading and running of tools. Tools are loaded from the src/tools/default_tools and src/tools/user_tools directories.
• Budget Management: The BudgetManager class in src/manager/budget_manager.py manages the resource and expense budgets for the project. It tracks the usage of resources and expenses and enforces budget limits.
• Model Integration: The project supports integration with various language models, including Ollama, Gemini, and Groq. The llm_models.py file defines abstract base classes for these integrations.

Usage

To use the project, follow these steps: 1. Install the required dependencies by running pip install -r requirements.txt. 2. Start the application by running python app.py. This will launch a web interface where you can interact with the agents and tools.

By default, on running python app.py, you would need to authenticate with Auth0. But, this can be overriden through the CLI argument --no-auth to skip authentication.

To use the project with additional tools and agents, you need to:

1. Configure the budget in src/tools/default_tools/agent_cost_manager.py.
2. Create tools and place them in the src/tools/default_tools or src/tools/user_tools directories.

Please note that by default, we do provide a lot of pre-defined tools and agents, so you may not need to create your own tools unless you have specific requirements.

Model Support

The project supports the following language model integrations: - Ollama: Local model management and invocation. - Gemini: Cloud-based model management and invocation from Google. - Groq: Cloud-based model management and invocation from Groq. - Lambda: Cloud-based model management and invocation from Lambda Labs.

Acknowledgements

We would like to thank Hugging Face, Groq and Lambda Labs for sponsoring this project and providing the necessary resources for development.

Citation

If you use this project in your research or applications, please cite it as follows: bibtex @misc{hashiruai, title={HASHIRU: Hierarchical Agent System for Hybrid Intelligent Resource Utilization}, author={Kunal Pai and Parth Shah and Harshil Patel}, year={2025}, eprint={2506.04255}, archivePrefix={arXiv}, primaryClass={cs.MA}, url={https://arxiv.org/abs/2506.04255}, }

Contributing

Contributions are welcome! Please submit pull requests with bug fixes, new features, or improvements to the documentation.