ChatCha: Decentralized AI Agent Platform

Overview

ChatCha is a decentralized AI agent platform where users can get their queries answered with data storage and verifiability provided by Filecoin/IPFS using Storacha. This project is built for the Hot AI Integrations, Hotter Storage & Spicy AI Collaboration with Storacha Data Storage Bounty, demonstrating decentralized data storage, real-time data sharing, and analysis.

What ChatCha Does

This project uses multiple specialized AI agents to answer user queries based on whether the user is asking for real-time latest data or information from various research papers. The multi-agent architecture ensures comprehensive and accurate responses.

🤖 Agents Used

1. Web Search Agent 🌐

• Searches the web in real-time
• Provides up-to-date information on current events and topics
• Stores search results and processing data in dedicated Storacha space

2. Scholar Agent 🎓

• Searches various articles and research papers
• Provides academic and research-backed information
• Maintains its own Storacha space for scholarly data storage

3. Analyzer Agent

• Takes output/data from both Web Search & Scholar Agents (i.e. from their spaces)
• Summarizes and synthesizes information from multiple sources
• Delivers concise, coherent final responses to users
• Stores analysis data in a separate Storacha space

ChatCha Workflow

Storacha Integrations 🔥

1. Unique Storacha Spaces

ChatCha creates 4 distinct Storacha Spaces:

• Web Agent Space: Where web agent stores its data
• Scholar Agent Space: Where scholar agent stores its data
• Main Space: Where combined data from both web & scholar agents is stored
• Analyzer Agent Space: Where analyzer agent stores its data

2. Storage & Retrieval of Artifacts

For each query, ChatCha stores four key artifacts:

• input – Original prompt provided to the agent
• output – Final result data
• conversation – Conversation between the user and agent
• metadata – Timestamp, model info, etc.

Have a reference from this URL: https://bafkreiaqhxxuum2vsayxdxskxj7adocwwgrscqgbp6ajjj26jzcx6gir2m.ipfs.w3s.link/

Code

Initializing Storacha Client

Creating Spaces

Uploading to spaces

Technologies Used

Feature	Stack
AI agent framework	Langchain (Groq)
AI Model	qwen-2.5-32b
Storage	Storacha
Frontend	Tailwind CSS + ReactJS

Demo Video

• Youtube Link ( 3 min ) : https://youtu.be/N1t-bySH3Q0?si=DH-JuBM4mbJWQhr4
• Youtube Link (Storacha Integration in Detail - 5 min) : https://youtu.be/XYBFjriG5Pg?si=9AyStUBkVHPgqUhS

Reference Links from Demo Video

• Web x Scholar CID : https://bafkreigju46wyofhzqkkefpuj5o2o3bpbe3rs2xx3jxg5ddxugfwmzhosm.ipfs.w3s.link/
• Analyzer CID : https://bafkreiehrpq6lqypdm3bwwjcjunezqyssphe7qa3xiw53f5g47z35av4gi.ipfs.w3s.link/

Links

Presentation : https://www.canva.com/design/DAGkU4BoctY/z2dbw7cOerUCVRhsgsCz8A/edit?utm_content=DAGkU4BoctY&utm_campaign=designshare&utm_medium=link2&utm_source=sharebutton

Project Structure

└── akashmundari-chatcha/
    ├── README.md
    ├── eslint.config.js
    ├── index.html
    ├── package.json
    ├── postcss.config.js
    ├── tailwind.config.js
    ├── vite.config.js
    ├── public/
    └── src/
        ├── App.jsx
        ├── index.css
        ├── main.jsx
        ├── assets/
        ├── chatcha/
        │   ├── mul.jsx
        │   └── sec.jsx
        └── components/
            └── test.jsx

Bounties Targeted

1. Hot AI Integrations, Hotter Storage

Requirements	Satisfied or Not	Notes
Create an account and signup for a plan	✅	Created an account on Storacha
Use an open-source AI agent framework	✅	Used Langchain along with open-source LLM
Create a space	✅	Space Created for storing data for various Agents
Upload Data 3+ items	✅	Data such as input, output, conversation and metadata are uploaded to Storacha
Download Data 3+ items	✅	User can be able to download input, output, conversation and metadata from Storacha
Documentation	✅	Provided as README.md
Project added to github discussions	✅	Yes
Video submission	✅	Yes
Frontend	✅	Tailwind CSS + React
Launch in 3 minutes or less	✅	Yes

2. Spicy AI Collaboration with Storacha Data Storage

Requirements	Satisfied or Not	Notes
Create an account and signup for a plan	✅	Created an account on Storacha
Use an open-source AI agent framework	✅	Used Langchain along with open-source LLM
Create a space	✅	Space Created for storing data for various Agents
Upload Data	✅	Data such as input, output, conversation and metadata are uploaded to Storacha
Download Data	✅	User can be able to download input, output, conversation and metadata from Storacha
Documentation	✅	Provided as README.md
Project added to github discussions	✅	Yes
Video submission	✅	Yes
Frontend	✅	Tailwind CSS + React
Launch in 3 minutes or less	✅	Yes

---