Skip to content

USE (USD stablecoin)#

Centralized USD-pegged stablecoins anchor most on-chain liquidity and activity. Their design brings censorship and regulatory risk. Refusing to interoperate isolates Ergo in a niche and reduces utility for regular users. This page lays out the case for a cost-efficient USD stablecoin for the Ergo ecosystem (USE), the core design, initial parameters and a practical roadmap for launch and operations.

Related docs: Use cases overview, Dexy


Summary#

  • Objective: deliver a cost-efficient, highly liquid USD-pegged asset (USE) that maximizes on-chain utility while preserving protocol security.

  • Positioning: a pragmatic asset that interoperates with broader crypto liquidity (USDT and peers) while keeping protocol-level safety strong and avoiding censorship-prone primitives at the platform layer.

  • Guiding principle: prioritize cost efficiency (low fees, scalable flows) without compromising security of protocol contracts.


Why USE on Ergo?#

  • Liquidity and onboarding: USD stablecoins function as primary on-ramps and liquidity anchors. Support for them raises accessibility for users and traders.

  • Ecosystem growth: Interoperability with the most liquid assets supports DEX volume, developer traction, and integrations.

  • Complementary design: A well-specified USE and its tooling can showcase Ergo’s advantages (privacy features, formal contracts) without forcing users to abandon familiar liquidity.


Design Choices and Recommendations#

  • Reserve model (1:1): adopt a conservative, transparent reserve approach that follows Dexy-style operational assumptions to reduce systemic risk.

  • Cost efficiency: select mechanisms that minimize fees and friction across composable flows. Dexy’s low-fee properties and Rosen Bridge integration can drive competitive costs.

  • Recommended fee targets:

    • Rosen and cross-chain bridges: enable custom fees as low as ~0.1% to remain competitive.
    • Dedicated v3 AMM instance (e.g., BSC): target 0.05% fees with a tighter price band (0.9–1.1) for stable-stable liquidity.
  • UI fees and fee caps: adopt a front-end fee that lifts after the protocol-wide collected fees for USE pass a configured threshold. The UI operator (example: C8) should set and enforce this policy for USE only.

  • Decimal precision: match major liquidity partners. Use 6 decimals as a minimum. This choice provides fine granularity (0.999999) and aligns with many stablecoins and exchanges. Avoid low-precision choices (for example, 2 decimals) that weaken tight stable-stable markets.

  • Supply considerations:

  • Supply size does not control peg mechanics. Choose a nominal supply that avoids awkward decimals and matches UX expectations.
  • Community discussions explored ranges from 50 m to very large nominal supplies. Very large nominal supplies (for example, 100 T) require 6 decimals to keep UX precise.

  • Token identifiers and governance tokens: prepare and publish canonical token IDs (pool NFT, refresh/update NFTs, oracle, reward, ballot) in initial docs and deployment manifests so trackers and UIs can follow them. Include example formats in release notes.

  • Oracles: select reliable price oracles and document decimal handling so USDT/USDC pairings behave as expected. Ensure oracle data follows the token’s decimal conventions.


Bank Contract: Stability and Security (Starting Properties)#

We aim for a stable peg and robust security under realistic adversarial conditions.

  • Reserves: 1:1 reserves in segregated, auditable addresses or multisig, with governance-controlled vaults.

  • Oracle design: multi-source inputs with signatures, aggregation, outlier rejection, and a pause control for anomalous readings.

  • Peg controls: conservative, auditable adjustments only; no privileged auto-minting.

  • Governance and upgradeability: minimal, explicit upgrade paths; emergency pause under multisig with time-locks to prevent unilateral actions.

  • Limits and caps: early per-block and per-wallet mint/redemption limits that reduce exposure to procedural or market errors.

  • Transparency: public reserve attestations and documented accounting procedures, with on-chain proofs where feasible.


Initial Parameter Suggestions (For Review)#

  • Decimals: 6.

  • Initial nominal supply: choose a number that avoids awkward UX. Community leaning: 250,000,000. A larger nominal supply also works when paired with 6 decimals.

  • Fee structure:

    • Protocol/bridge configurable fee floor at 0.1%, with operator flexibility for competitive rates.
    • Target 0.05% for dedicated stable-stable AMM pools.
    • UI fee waiver after a defined cumulative-fees threshold for USE (threshold set by the UI operator).
  • Oracle cadence: conservative configuration with minute-level updates plus aggregation and outlier rejection.

  • Mint/redemption onboarding: start with off-chain KYC/AML for minting partners if counterparties require it. Publish clear KYC policies.


Operational Notes and Trackers#

  • Pre-launch tokens: no tokens exist until mint. Indexers and off-chain bots should prepare to track and register token IDs once minted. Trackers should ingest token metadata when available.

  • Publication: publish all token IDs and NFT IDs for pools, refresh, update, oracle, reward, and ballot before or at launch so UIs and indexers can enable support.


Roadmap (Proposal)#

  • Finalize high-level design and governance model (multisig, time-locks, upgrade policy).

  • Decide decimals and nominal supply (community vote or dev consensus).

  • Implement bank contract and complete a security review and audit.

  • Implement oracle adapters and aggregator logic; test on testnet.

  • Prepare Rosen Bridge integration and expose custom fee options.

  • Deploy initial tokens on testnet; publish token IDs.

  • Prepare UI integrations, indexers, and off-chain bot updates.

  • Launch mainnet mint with coordinated liquidity onboarding (DEX pools, AMM instances).

  • Continue audits, reserve transparency, and fee tuning based on activity.


Suggested Role Assignments (Initial)#

  • Protocol and bridge integration: Rosen maintainers coordinate custom lower-end fee settings.

  • AMM and liquidity engineering: Dexy and AMM operators define tight stable-stable pools; consider a v3 instance on BSC for cross-chain liquidity tests.

  • UI fee policy and front-end: C8 defines the threshold and implements fee-lifting for USE.

  • Token spec and decimals: Kushti documents implementation details and minting parameters.

  • Audit and security: independent auditors and core dev reviewers oversee scope and fixes.

  • Community coordination: one or two community managers publish token IDs and coordinate with exchanges and trackers.


Open Questions Before Launch#

  • Exact nominal total supply and decimal decision
    Proposed: 6 decimals, nominal supply at 250 m or a different nominal with 6 decimals.

  • Final governance signers and time-lock parameters.

  • Audit scope and dates.

  • Oracle providers and aggregator logic.


Notes From Community Discussions#

  • Broad support exists for 6 decimals to match USDT and support tight pricing.

  • Dexy’s transparent 1:1 reserve approach and cost focus fit the target UX for USE.

  • Very low fees attract flow; security and reserve integrity remain the higher priority.

  • A UI fee-waiver after a collected-fees threshold can strengthen UX for USE.


Next Steps#

  1. Confirm the decimals and nominal supply choice and record the decision.

  2. Start the bank contract specification and nominate multisig signers.

  3. Produce a detailed technical spec for the bank contract and AMM parameters for audit.

  4. Assign owners for roadmap items and set timelines.