Project_AB
On-chain AI Code Audit & Provenance Stamp
Created on 10th August 2025
•
Project_AB
On-chain AI Code Audit & Provenance Stamp
The problem Project_AB solves
Current code audit and provenance verification processes are either manual, expensive, or require trusting a central service.
This project automates the process of analyzing a GitHub repository, generating a verifiable report, storing it on IPFS, and recording the IPFS CID on-chain.
People can use it to:
Provide tamper-proof audit reports for open-source projects.
Give verifiable proof-of-code for hackathon submissions or grant applications.
Ensure long-term availability of code state at a given commit.
Build trust with users, investors, or the community by providing immutable, on-chain references to their work.
Challenges I ran into
One of the main challenges was designing a seamless integration between the GitHub repository analysis process, IPFS storage, and on-chain recording.
We needed to ensure that:
The frontend stayed responsive while polling job status in real time.
The IPFS upload and blockchain transaction steps were reliable and resilient to temporary failures.
The user experience was simple enough for non-technical users, even though the underlying process involves complex decentralized storage and smart contract interactions.
Balancing these technical requirements while keeping the UI clean and intuitive required multiple iterations and careful architectural decisions.
Smart Contract Address(es)
Other Onchain Proof
All reports are recorded on-chain with their IPFS CID. Example transaction:
https://ipfs.io/ipfs/bafybeidwul7iqelgg426yjq5takqemrthxrrq3fvk22fhcmola2ils4cla/
Technologies used