https://github.com/mathyscogne/hackathon_kiln
Solution for Hackathon Kiln 2025 | AI-powered assistant that both educates on blockchain concepts and can help advanced users quickly retrieve precise, real-time data using Kiln's API
https://github.com/mathyscogne/hackathon_kiln
blockchain chatbot ia
Last synced: 6 months ago
JSON representation
Solution for Hackathon Kiln 2025 | AI-powered assistant that both educates on blockchain concepts and can help advanced users quickly retrieve precise, real-time data using Kiln's API
- Host: GitHub
- URL: https://github.com/mathyscogne/hackathon_kiln
- Owner: MathysCogne
- Created: 2025-02-01T08:28:37.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-02-04T14:39:12.000Z (8 months ago)
- Last Synced: 2025-03-25T14:05:26.113Z (7 months ago)
- Topics: blockchain, chatbot, ia
- Language: Python
- Homepage:
- Size: 18.9 MB
- Stars: 6
- Watchers: 1
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Hackathon - Kiln 🐧
2nd Place WINNERS of the Kiln Connect category! 🏆
This project was developed as a solution for a hackathon organized by Kiln, focusing on innovation and data accessibility in staking and blockchain technology 🏆
## 🚀 Project Objective
The goal is to provide an AI-powered assistant that both educates on blockchain concepts and can help advanced users quickly retrieve precise, real-time data using Kiln's API. Acting as a bridge to the blockchain, the AI makes information more accessible, and actionable.
# Slide presentation
## 🔎 Sentiment analysis
The first step is to have the user's prompt go through a simple sentiment analysis using a python script, allowing it to narrow down which data is necessary and from there make the right API call to retrieve it. In a fully fledged version, a more thorough sentiment analysis using modified [Afinn](https://github.com/fnielsen/afinn) and allowing it to understand actionable words to enable it to make actions (such as handling: transactions, withdrawing, staking on behalf of the user).
## 📡 Integration with Kiln
The aim of this project was to demonstrate a functional proof of concept, for this we decided to limit the data received through Kiln's API to what we thought would be necessary for a MVP.
For a fully operational version, all Kiln API endpoints should be integrated to maximize the chatbot's capabilities.
The AI is fed real-time data from Kiln's API to provide accurate responses. Currently supported endpoints:
- `Get_Stakes`: Information on validators and associated wallets
- `Get_Reward`: Reward details for a given wallet or validator
- `Get_ETH_Stats`: Global Ethereum network statistics
- `Get_Kiln_Stats`: Kiln-specific activity data on Ethereum
- `Get_Transaction_Status`: Status of an Ethereum transaction
## 🔮 Future Improvements
- 🛠 **Blockchain Actions**: Integration of Kiln's UI widget for direct user actions
- 📡 **Extended Kiln API Support**: Add new API endpoints for broader information coverage
- 🤖 **AI Model Enhancement**: Use a blockchain-trained LLM model locally, to improve response quality, privacy and compliance
## 🙏 Acknowledgments
Special thanks to Kiln for organizing the hackathon!
