Skip to main content

IDOL / Arrakis — Improving DEX / On-chain Liquidity

Forum thread: https://forum.livepeer.org/t/pre-proposal-idol-improving-dex-onchain-liquidity/3151

Summary​

This dossier evaluates a DEX-liquidity-focused proposal (often discussed alongside Arrakis-style vault management) and keeps the analysis reproducible:

  • what the mechanism is (protocol-owned or incentivized liquidity)
  • what it plausibly changes for delegators (conversion friction, slippage, on-chain UX)
  • what can go wrong (custody/admin risk, IL/LVR, wash trading, “looks good in a dashboard” failure modes)

Quick start (reproduce)​

From this directory:

  • Fetch forum thread snapshot: python3 scripts/fetch_forum_topic.py
  • Generate onchain slippage table (Uniswap v3 pool on Arbitrum): python3 scripts/onchain_slippage_univ3.py --include-tick-depth
  • Pool volume/fees from swap logs: python3 scripts/univ3_swap_analytics.py --pool <POOL> --token0-coingecko-id <id> --token1-coingecko-id <id> --days 1
  • Address-level swap concentration (wash-trade surface): python3 scripts/univ3_swap_address_analytics.py --pool <POOL> --token0-coingecko-id <id> --token1-coingecko-id <id> --days 30
  • “Reversion after swaps” analysis: python3 scripts/univ3_reversion_analysis.py --pool <POOL> --token0-coingecko-id <id> --token1-coingecko-id <id> --days 1 --include-time
  • Chunked vs single trade best-case comparison: python3 scripts/chunked_trade_analysis.py --pool <POOL> --token0-coingecko-id <id> --token1-coingecko-id <id> --total-usd 25000 --chunk-usd 1000
  • Run IL + breakeven volume models: python3 scripts/il_models.py
  • Download + summarize relevant Arrakis audit PDFs: python3 scripts/audit_summaries.py
  • Run cross-pool case studies (slippage + volume/fees): python3 scripts/run_case_studies.py

Outputs land in outputs/ and cached downloads land in data/.

Notes​

  • The proposal text references “Uniswap v4”, but the existing LPT/WETH pool used in practice is Uniswap v3 on Arbitrum.
  • Re-run scripts before sharing numbers; this analysis depends on live onchain state and current prices.

Hardhat sandbox​

Toy simulations live in sim-hardhat/ (ignored by git via node_modules/ etc). Example:

  • cd sim-hardhat && npm run simulate:wash
  • Forked slippage sanity check (adds 50/50 liquidity on an Arbitrum fork and re-quotes):
    • cd sim-hardhat && SIM_OUT=../outputs/fork.json SIM_TOTAL_USD=782000 SIM_RANGE_MULT=60 npx hardhat run --network hardhat scripts/forkLiquidityImpactSim.js

Outputs and data/ are committed in this repo for “zero-friction browsing” (stakeholders can read results without rerunning scripts).