Usage Guide

Architecture & Workflow

• User Inputs: The user provides the recipient address and the ETH amount in the AI-powered UX.

• Transaction Method Selection: The system determines whether to process a transfer within the same rollup or bridge across rollups.

• Processing: The transaction is handled through the rollup protocol.

• Settlement Verification: Espresso Systems ensures transaction finality.

• Completion: The transaction is confirmed, and the user receives a notification.

---

Bridging Assets

1. Connect your wallet using the "Connect" button

2. Select source and destination chains from the dropdown menus

3. Enter the amount of ETH to bridge

4. Click "Bridge" and confirm the transaction in your wallet

5. Monitor transaction status in real-time

Interacting with Luwak AI

• Click the menu "Chat" in the navbar

• Ask questions about Espresso technology, bridges, or current prices

• Receive guidance throughout your bridging process

• Experience the unique cyberpunk raccoon persona

---

For Developers

How to Give Knowledge to Luwak AI Agent

Luwak AI uses a Retrieval-Augmented Generation (RAG) system to access and process information. You can easily expand its knowledge base by following these steps:

1. Prepare Documentation Files:

   • Convert your documentation to PDF format

   • Place the PDF files in the embedding_docs/input/ directory

2. Process the Files:

   • Run the application locally:

     yarn dev

   • Or in production:

     yarn build && yarn start

3. Generate Embeddings:

   • Access the embedding creation endpoint:

     http://localhost:3000/api/embedding/create

   • The system will automatically:

     • Convert PDFs to text files (stored in embedding_docs/output/)

     • Create vector embeddings from the text

     • Store these embeddings in the database for AI retrieval

4. Verification:

   • Check the terminal output for processing status

   • Test the AI with questions related to your new content

   • The knowledge is now available for Luwak AI to use in responses

Note: The embedding process may take several minutes depending on the size and number of documents being processed.

Custom Configurations

• Modify chain configurations in /lib/chains.js

• Adjust Luwak AI's persona settings in /app/api/agent/characters.js

• Extend supported tokens in /lib/contracts.js