Deployments
Where AquaQoS has run, which environment each receipt comes from, and how to recheck it.
| Environment | Label | What it is |
|---|---|---|
| Benchmark report | Recorded | 72 fresh local EVM fixtures with retained receipts |
/live/ workspace | Live local | A fresh isolated Cancun EVM with mock tokens, per session |
| Local Ethereum fork | Local fork | Authenticated official Aqua and real DAI/WETH on a local copy of block 25,948,160 |
| Ethereum Sepolia | Public testnet | A public testnet deployment with valueless demo tokens |
| Mainnet | Planned | None. No mainnet deployment exists |
Ethereum Sepolia
Chain ID 11155111, verified 2026-09-11 from clean source 17a6b99. All four custom contracts
are exact Sourcify creation and runtime matches.
Public testnet, not production
The demo tokens are valueless and owner-mintable. This proves deployed behavior on a public chain; it is not a production deployment or a security certification.
| Contract | Address | Source |
|---|---|---|
| AquaQoS router | 0xE2EE…5496 | Sourcify exact match |
| AquaQoS vault | 0x22a3…C252 | Sourcify exact match |
| Demo token aqA | 0x5f0a…F5DC | Sourcify exact match |
| Demo token aqB | 0x6242…374a | Sourcify exact match |
The router uses the official Aqua deployment at
0x4999…6d31.
Its 6,251-byte runtime matches the separately authenticated Ethereum deployment.
Two strategies each hold a 500 aqA guarantee over shared vault inventory:
| Action | Result | Transaction |
|---|---|---|
| Strategy 1 requests 600 aqA | Reverted with InsufficientCapacity; state and logs unchanged | 0x5ad5…7bbd |
| Strategy 1 requests 500 aqA | Filled; three ERC-20 Transfer logs | 0x49f7…3c42 |
| Strategy 2 requests 501 aqA | Reverted with InsufficientCapacity; state and logs unchanged | 0x6a76…c61d |
| Strategy 2 requests 500 aqA | Filled; three ERC-20 Transfer logs | 0x1696…e968 |
Replenish strategy 1 through Aqua.push | Capacity restored | 0x0c09…6b5e |
| Strategy 1 requests the restored 500 aqA | Filled | 0x7c99…9ff2 |
| Reverse-direction request | Filled 100 aqB output | 0x78dd…6309 |
The complete run retains 22 public transactions, including setup, and cost 0.012706443810287423 Sepolia ETH in total.
Local Ethereum fork
The fork proof runs AquaQoS against Ethereum state copied into an in-process Hardhat EVM. Deployments, funding and swaps are all local; there is no Ethereum inclusion and no public explorer link for these receipts.
- Authentication. The official Aqua address listed for Ethereum hosts an older
AquaRouterbuild. Its retained compiler input recompiles to the exact 6,251-byte runtime at block 25,948,160, including metadata. - Real tokens. DAI is funded from an impersonated account in the local copy; WETH through its actual deposit function. No token storage is patched and no mock token is used.
- Same story. Raw Aqua fills 0.6, then the sibling's 0.5 fails for inventory shortage. AquaQoS rejects the 0.6 (1.1 required), admits 0.5 for each strategy, rejects 0.501, and then replenishes, reverses and exits to an empty vault.
Clean source de8c609 produced both retained reports with dirty=false: 45 local and 43 fork
transactions, each with eight compressed swap traces. The checker rejects seven deliberately
corrupted reports per environment.
Recheck it yourself
Re-query Sepolia and Sourcify rather than trusting the committed report:
pnpm check:sepoliaThe checker binds committed source hashes, the authenticated Aqua runtime, deployed runtime hashes, receipts, exact Sourcify matches, rollback states, transfer logs, final balances and allowances. See Reproduce for the fork run.