ClearTxn

This project has a user-friendly interface that anyone can visit to input any transaction. After the user submits a transaction, the system processes it and provides a straightforward explanation of what the transaction did. This way, users can easily understand the details and effects of their transactions without needing any technical knowledge.

When a user inputs a transaction on the front-end, it is sent to the back-end for processing. The back-end uses the starknet_simulateTransactions method to simulate the transaction. This simulation provides various useful data points, including the different function calls and any emitted events.

The back-end extracts these function calls and events, then uses the Voyager API to fetch the ABIs (Application Binary Interfaces) for the class hashes. From these ABIs, it retrieves the function and event names. For certain events, additional information is extracted as well.

All the gathered information is then formatted into a prompt for a Large Language Model (LLM). The LLM is tasked with translating this technical transaction data into a human-readable explanation for the user.

Finally, the back-end sends this explanation back to the front-end, where it is displayed to the user.