PrivyPass

PrivyPass is an AI privacy credential copilot that helps users prove eligibility with Midnight-powered zero-knowledge proofs—without revealing unnecessary personal or financial data.

Comment

Inspiration

What it does

How we built it

Challenges we ran int## Inspiration

Imagine applying online for a scholarship, loan, student housing, or another eligibility-based service.

The platform usually does not need to see a user's complete financial profile. It only needs to know whether the user meets a specific requirement.

However, traditional applications often require people to upload sensitive documents and reveal much more information than necessary.

We built PrivyPass for people who need to prove something without exposing everything.

PrivyPass is an AI-first privacy credential copilot created for the Midnight Hackathon. It combines conversational AI, selective disclosure, and zero-knowledge proofs to make private digital verification easier to understand and use.

What it does

PrivyPass turns a user's goal into a privacy-preserving verification workflow.

A user can describe what they are applying for, and the AI copilot helps identify what needs to be proven. In our current MVP, PrivyPass focuses on financial eligibility.

The workflow is:

  1. The user confirms a public eligibility requirement.
  2. PrivyPass creates a reusable credential.
  3. The application generates a Midnight-backed zero-knowledge proof.
  4. A verifier checks whether the requirement is satisfied.
  5. The verifier receives only the result, not the user's complete private financial information.

For example, a user can prove that their income meets a required threshold without revealing their exact income.

PrivyPass also includes credential management, proof history, reusable templates, and a community area where verification use cases can be explored.

How we built it

The frontend was built with React, TypeScript, and Vite. It provides an AI copilot interface, credential management, proof generation, verification, templates, and community experiences.

The backend was built with Node.js, Express, and TypeScript. It is organized into modules for AI analysis, credential creation, proof generation, verification, application workflows, and HTTP APIs.

We used the OpenAI API to translate a user's natural-language request into a structured proof requirement.

For the privacy layer, we integrated Midnight, including:

  • Compact smart contract logic
  • Midnight node
  • Midnight indexer
  • Midnight proof server
  • Midnight.js
  • Zero-knowledge proof generation
  • On-chain public state and proof verification

Our financial eligibility contract evaluates whether a private income value satisfies a public threshold. The private value is not revealed to the verifier.

The complete application flow connects the AI requirement, credential, Midnight proof, proof ID, public state, and final verification result.

Challenges we faced

One of the biggest challenges was connecting the complete Midnight development environment.

The node, indexer, and proof server all had to be available and correctly configured before the backend could generate a real proof. We also needed to distinguish genuine Midnight proof generation from fallback or unavailable states.

Another challenge was connecting multiple layers of the application:

  • Natural-language AI input
  • Structured eligibility requirements
  • Credential creation
  • Compact contract execution
  • Proof generation
  • Public state
  • Verification
  • Frontend status updates

We also had to design the product so that complex privacy technology felt understandable to users who may not know anything about zero-knowledge proofs.

Accomplishments that we are proud of

We are proud that PrivyPass demonstrates a complete privacy credential workflow rather than only presenting a visual prototype.

During the demonstration:

  • The Midnight node, indexer, and proof server are running.
  • The backend confirms that Midnight is configured.
  • Proof generation produces a real proof ID and public state.
  • Verification returns a real validity and eligibility result.
  • Private financial information is not exposed to the verifier.

We are also proud of creating an AI-first interface that turns a technical zero-knowledge proof process into a guided conversation.

What we learned

We learned that privacy technology becomes much more useful when users do not have to understand every technical detail behind it.

Midnight provides the privacy infrastructure, while the AI copilot helps users understand what they need to prove and guides them through the workflow.

We also learned the importance of clearly separating:

  • Private user data
  • Public verification requirements
  • Proof generation
  • Public state
  • Verification results

This separation helped us design both the contract and the user experience more carefully.

What's next for PrivyPass

Our next steps include:

  • Supporting additional credential types beyond financial eligibility
  • Adding scholarship, housing, employment, age, and membership templates
  • Improving document extraction and credential creation
  • Adding signed and portable proof certificates
  • Expanding reusable community templates
  • Supporting organization-created verification requirements
  • Improving persistent credential and conversation storage
  • Deploying a production-ready Midnight environment

Our long-term goal is to make privacy-preserving credentials accessible to everyday users and organizations.

PrivyPass doesn't ask users to reveal more. It helps them prove exactly what matters.o

Accomplishments that we're proud of

What we learned

What's next for PrivyPass

Built With

Share this project:

Updates

Submission history