Skip to main content

​
The Privacy Paradox

In traditional finance, trust requires transparency. A bank cannot lend to you unless they see your bank statements. In Web3, transparency is absolute—everything on-chain is public. This is a dealbreaker for businesses who treat their financial data as a trade secret. Manteia solves this with the Blind Oracle.

​
How It Works

We use Zero-Knowledge Proofs (ZK-SNARKs) to separate the verification of data from the data itself.

Circuit: revenue_check.circom

The core of our privacy layer. This circuit accepts private inputs (your revenue history) and outputs a public proof.

​
The Workflow

  1. Local Execution: The Manteia Client runs locally in your browser. It fetches data directly from Stripe/Shopify APIs via HTTPS.
  2. Witness Generation: The client inputs this data into the revenue_check circuit.
  3. Proof Generation: The circuit checks the data against Manteia’s underwriting rules (e.g., “Is 6-month average > $10k?”). It generates a cryptographic proof.
  4. On-Chain Verification: You submit only the proof to the Manteia Smart Contract. The contract verifies the math. If correct, it accepts the risk score without ever seeing the raw dollar amounts or customer names.

​
Data Integrity

How do we know the user didn’t fake the API response?
  • TLSNotary / MPC: We utilize Multi-Party Computation methods (similar to DECO or TLSNotary) to attest that the data originated from the specific server (e.g., api.stripe.com) and was not tampered with in transit.

​
Output

The “Blind Oracle” produces three public signals:
  1. Risk Score: (0-100) Computed based on volatility, churn, and growth.
  2. Borrow Limit: The maximum amount the protocol allows this entity to borrow.
  3. Integrity Hash: A hash of the dataset for future dispute resolution (optional).