Inspiration The inspiration for FrostDAO came from a critical flaw in current DAO and multisig tooling: "The Flat Governance Problem."
In the real world, organizations are hierarchical. A CEO has more authority than an intern; a parent has more authority than a child. However, current blockchain tools (like standard Multisigs or TSS) treat every key holder as equals. In a "3-of-5" setup, three junior employees (or compromised keys) can collude to drain a treasury, bypassing the Founder entirely. We call this the "Janitor Attack."
We realized that "Code is Law" shouldn't mean "All Keys Are Equal." We wanted to build a system that respects the Chain of Command at the cryptographic level, ensuring that critical actions always require executive approval, no matter how many lower-ranked members agree.
What it does FrostDAO is a Hierarchical Threshold Signature Scheme (HTSS) management platform. Unlike traditional multisigs that rely on a simple headcount (e.g., "any 3 people"), FrostDAO enforces Rank-Based Authorization.
It allows organizations to structure permissions where:
Rank 0 (e.g., CEO/Founders): Must be present for any transaction to be valid.
Rank 1 (e.g., Managers): Can sign, but cannot execute without a Rank 0 member.
Rank 2 (e.g., Community/Interns): Can participate in voting, but hold no execution power on their own.
If a group of Rank 2 members attempts to sign a transaction without a Rank 0 key, the cryptographic signature is mathematically invalid and rejected by the network. It brings true Sovereign Hierarchy to decentralized organizations.
How we built it The core innovation of FrostDAO lies in replacing the standard mathematics of Secret Sharing.
From Lagrange to Birkhoff: Traditional Shamir’s Secret Sharing uses Lagrange Interpolation, which connects a series of points that are all treated equally. We replaced this with Birkhoff Interpolation.
Derivative Constraints: By using Birkhoff, we assign different mathematical properties to keys. "Rank 0" keys provide standard positional values, while lower-ranked keys provide "derivative" values (or slope constraints).
The Reconstruction Logic: We built a custom polynomial reconstruction algorithm that solves this "Lacunary Interpolation" problem. The private key can only be reconstructed if the specific mathematical requirements of the hierarchy (the presence of Rank 0) are met.
Challenges we ran into The math was incredibly difficult.
The "Unsolvable Matrix" Problem: Unlike Lagrange interpolation (which is always solvable), Birkhoff interpolation can sometimes result in singular matrices where no solution exists. We had to carefully design our key generation parameters to ensure the system is always stable.
Lacunary Gaps: Handling "gaps" in data (where we know the derivative but not the position) required complex matrix operations that are not found in standard crypto libraries.
UX vs. Security: Abstracting this complex math into a simple "Family" or "Company" dashboard was a design challenge.
Accomplishments that we're proud of We fixed the "Janitor Attack": We successfully demonstrated a scenario where a majority of signers (Rank 1 & 2) tried to move funds and failed because the Rank 0 key was missing.
Mathematical Innovation: We successfully implemented a working proof-of-concept for Birkhoff-based Secret Sharing, which is rarely seen in production.
Versatility: We proved that this model works for diverse use cases—from Corporate Treasuries (CEO approval) to Family Trusts (Parental approval).
What we learned Math is powerful: We learned that governance rules don't need to be written in complex Smart Contracts (which can be hacked); they can be baked directly into the cryptography itself.
The gap in tooling: We realized how underserved "structured" organizations are in Web3. Most tools assume a flat, democratic structure that simply doesn't work for businesses or families.
What's next for FrostDAO Optimized Key Generation: We plan to rewrite the core interpolation logic in Rust/WASM to speed up the signing process.
Dynamic Hierarchies: Currently, ranks are set at creation. We want to enable "promotions" (upgrading a Rank 1 key to Rank 0) without generating a new wallet.
Integration: We aim to wrap this logic into an SDK so that existing wallets (like Gnosis Safe or Metamask Snap) can offer "Hierarchical Mode" as a plugin.
Log in or sign up for Devpost to join the conversation.