add test report#2942
Merged
sparrowDom merged 18 commits intoJul 21, 2026
Merged
Conversation
sparrowDom
requested review from
clement-ux,
naddison36 and
shahthepro
as code owners
July 16, 2026 21:33
…ation/test-report
…ation/test-report
clement-ux
reviewed
Jul 21, 2026
Comment on lines
+13
to
+14
| Only `mainnet`, `base`, `arbitrum`, `hyperevm` have fork endpoints in `foundry.toml` — **`sonic` and | ||
| `plume` are not configured**, so anything needing those forks is blocked. |
Collaborator
There was a problem hiding this comment.
We don't plan to support sonic and plume anymore. So I'll not add these chain endpoints.
| | 2 | Sonic (OSVault auth, wOS config, SwapX yield) | vault-general, token-wrapped | ~7 | No `sonic` endpoint in `foundry.toml` | Add `sonic = "${SONIC_PROVIDER_URL}"` so these can be ported, or skip? | **Skip** — Sonic is winding down; all funds sit in the vault buffer and the strategies no longer hold any assets, so there is nothing to smoke-test. | | ||
| | 3 | OUSD Rebalancer suite | rebalancer | 109 | Off-chain JS + GraphQL (`utils/rebalancer.js`) — not Solidity, can't run in forge | Re-home as a standalone JS runner (mocha/vitest), or accept loss once Hardhat CI is gone? | | | ||
| | 4 | decode-origin-nonce | crosschain | ~5 | Off-chain JS decoder (`tasks/crossChain.js`) | Same as #3 — JS runner or drop? | | | ||
| | 5 | Algebra / Hydrex AMO (`StableSwapAMMStrategy`) | strat-algebra-amo | 69 | Contract exists but **not deployed anywhere**; Hydrex was withdrawn; the Sonic SwapX variant is already covered | Deprecated (skip), or coming to Base (unit tests now, fork later)? | | |
Collaborator
There was a problem hiding this comment.
We won't use it anymore, so I didn't port them and don't plan to do it.
| | 7 | `RebalancerModule` full unit suite | safe-modules | 46 | Contract exists, **not yet deployed**; unit-testable now (like `AutoWithdrawalModule`) | Implement the full unit suite now? (No hard blocker — confirm priority given it's not deployed.) | | | ||
| | 8 | Base `SuperOETHHarvester` | vault-multichain | ~7 (+8 retired `it.skip`) | No Foundry harvester test infra; the `harvestAndSwap` cases are retired | Build harvester infra (whitelist / dripper / `harvestAndTransfer`)? Confirm the retired `harvestAndSwap` `it.skip` cases are dropped. | **Skip** — retired `harvestAndSwap` `it.skip` cases dropped. | | ||
| | 9 | Base/HyperEVM Timelock governance | zapper-gov-hacks | 2 | `GovHelper` implements only the mainnet GovernorSix flow | Extend `GovHelper` for `TimelockController` (Base + HyperEVM), or defer? | | | ||
| | 10 | Legacy OUSD migration-state tests (altCPT ≠ 1e18) | token-ousd | ~4 | Require `vm.store`-forging legacy account state | Implement with state-forging, or defer? | | |
Collaborator
There was a problem hiding this comment.
Very old legacy works. I don't think we need to keep tests there.
| | 4 | decode-origin-nonce | crosschain | ~5 | Off-chain JS decoder (`tasks/crossChain.js`) | Same as #3 — JS runner or drop? | | | ||
| | 5 | Algebra / Hydrex AMO (`StableSwapAMMStrategy`) | strat-algebra-amo | 69 | Contract exists but **not deployed anywhere**; Hydrex was withdrawn; the Sonic SwapX variant is already covered | Deprecated (skip), or coming to Base (unit tests now, fork later)? | | | ||
| | 6 | Legacy `NativeStakingSSVStrategy` | strat-native-ssv | 42 | Legacy strategy, superseded by `CompoundingStakingSSVStrategy` (already unit-tested) | Port the legacy suite, or retire it (skip / minimal smoke only)? | **Skip** — legacy strategy retired, superseded by CompoundingStakingSSVStrategy. | | ||
| | 7 | `RebalancerModule` full unit suite | safe-modules | 46 | Contract exists, **not yet deployed**; unit-testable now (like `AutoWithdrawalModule`) | Implement the full unit suite now? (No hard blocker — confirm priority given it's not deployed.) | | |
Collaborator
There was a problem hiding this comment.
Never deployed, not used and not planned to. We don't need to add tests there.
| | 8 | Base `SuperOETHHarvester` | vault-multichain | ~7 (+8 retired `it.skip`) | No Foundry harvester test infra; the `harvestAndSwap` cases are retired | Build harvester infra (whitelist / dripper / `harvestAndTransfer`)? Confirm the retired `harvestAndSwap` `it.skip` cases are dropped. | **Skip** — retired `harvestAndSwap` `it.skip` cases dropped. | | ||
| | 9 | Base/HyperEVM Timelock governance | zapper-gov-hacks | 2 | `GovHelper` implements only the mainnet GovernorSix flow | Extend `GovHelper` for `TimelockController` (Base + HyperEVM), or defer? | | | ||
| | 10 | Legacy OUSD migration-state tests (altCPT ≠ 1e18) | token-ousd | ~4 | Require `vm.store`-forging legacy account state | Implement with state-forging, or defer? | | | ||
| | 11 | Whale `withdrawAllFromStrategies` on real strategies | vault-oeth, vault-general | 3 | Heavy fork test that unwinds real deployed mainnet strategies via the timelock | Implement the heavy end-to-end fork test, or defer? | | |
Collaborator
There was a problem hiding this comment.
I don’t think this heavy end-to-end test is worth porting.
The behavior is already covered compositionally: unit tests exercise withdrawAllFromStrategies() and the withdrawal queue, while fork/smoke tests exercise withdrawAll() on the real strategies individually.
Running the full flow against live mainnet state would also require significant state manipulation, for example, ensuring enough Morpho liquidity and funding the vault so a whale withdrawal can be claimed.
Once we rely on those cheats, the test provides little additional confidence over the existing focused unit and fork tests, where these conditions are already controlled and tested with clearer assertions.
|
|
||
| | Item | Area | Why deferred | | ||
| |------|------|--------------| | ||
| | Beacon exact SSZ roots + `0x01` validator vector | beacon | Need byte-exact input vectors / a real valid proof from the Hardhat fixtures to reproduce the known constants (`0x5b449f…`, `0xc27ca5…`); the current tests only assert `!= bytes32(0)` | |
Collaborator
There was a problem hiding this comment.
I’ll try to port the exact SSZ root assertions for merkleizeSignature and merkleizePendingDeposit. Unlike the full 21-validator scenarios, these are small, deterministic unit vectors that require no fork or storage manipulation, and they provide meaningful protection against field-
ordering, padding, and endianness errors.
I do not plan to port the large 0x01 validator proof unless it turns out to be straightforward and adds coverage beyond the existing real-proof fork tests.
| | Item | Area | Why deferred | | ||
| |------|------|--------------| | ||
| | Beacon exact SSZ roots + `0x01` validator vector | beacon | Need byte-exact input vectors / a real valid proof from the Hardhat fixtures to reproduce the known constants (`0x5b449f…`, `0xc27ca5…`); the current tests only assert `!= bytes32(0)` | | ||
| | Curve AMO front-running strengthen | strat-curve-amo-mainnet | Needs deep reading of the existing fork test + exact protocol-profit accounting (`Δ totalValue − Δ totalSupply`), plus the OUSD/USDC 6-decimal variant | |
Collaborator
There was a problem hiding this comment.
Good call.
I’ll strengthen the existing Foundry front-running tests. The current totalValue >= totalSupply assertion can hide a loss when the Vault already has a surplus. Tracking Δ totalValue − Δ totalSupply directly provides a stronger regression check that the attack and subsequent unwind do not extract value from the protocol.
I’ll also cover the OUSD/USDC variant if the existing fixture can support it cleanly.
| |------|------|--------------| | ||
| | Beacon exact SSZ roots + `0x01` validator vector | beacon | Need byte-exact input vectors / a real valid proof from the Hardhat fixtures to reproduce the known constants (`0x5b449f…`, `0xc27ca5…`); the current tests only assert `!= bytes32(0)` | | ||
| | Curve AMO front-running strengthen | strat-curve-amo-mainnet | Needs deep reading of the existing fork test + exact protocol-profit accounting (`Δ totalValue − Δ totalSupply`), plus the OUSD/USDC 6-decimal variant | | ||
| | Base AMO exact params (`allowedWethShareInterval`, harvester) | strat-base-amo | Exact values are governance-tunable — need on-chain confirmation before asserting | |
Collaborator
There was a problem hiding this comment.
I don’t plan to add exact assertions for allowedWethShareInterval or the harvester address. The interval is governance-tunable, and its validity and behavior are already covered by unit and smoke tests. The harvester is also live configuration that can change through governance. Pinning either value would make the tests brittle; exact production configuration is better handled by deployment validation and off-chain monitoring.
| | Beacon exact SSZ roots + `0x01` validator vector | beacon | Need byte-exact input vectors / a real valid proof from the Hardhat fixtures to reproduce the known constants (`0x5b449f…`, `0xc27ca5…`); the current tests only assert `!= bytes32(0)` | | ||
| | Curve AMO front-running strengthen | strat-curve-amo-mainnet | Needs deep reading of the existing fork test + exact protocol-profit accounting (`Δ totalValue − Δ totalSupply`), plus the OUSD/USDC 6-decimal variant | | ||
| | Base AMO exact params (`allowedWethShareInterval`, harvester) | strat-base-amo | Exact values are governance-tunable — need on-chain confirmation before asserting | | ||
| | token-ousd account-type transfers/mints | token-ousd | Need a multi-account-type fixture (rebasing / non-rebasing / delegation-target) | |
Collaborator
There was a problem hiding this comment.
Good call.
I don’t think the full Cartesian account-type matrix needs to be ported. Foundry already covers the main rebasing/non-rebasing transitions, automatic account migration, legacy CPT normalization, yield delegation, and supply invariants.
There is still value in adding a few focused cases for transfers and mints involving YieldDelegationSource and YieldDelegationTarget, since those states have dedicated accounting branches. I’ll cover those missing directions directly rather than rebuilding the large and repetitive Hardhat fixture matrix.
| | Curve AMO front-running strengthen | strat-curve-amo-mainnet | Needs deep reading of the existing fork test + exact protocol-profit accounting (`Δ totalValue − Δ totalSupply`), plus the OUSD/USDC 6-decimal variant | | ||
| | Base AMO exact params (`allowedWethShareInterval`, harvester) | strat-base-amo | Exact values are governance-tunable — need on-chain confirmation before asserting | | ||
| | token-ousd account-type transfers/mints | token-ousd | Need a multi-account-type fixture (rebasing / non-rebasing / delegation-target) | | ||
| | strat-behaviour `setHarvester` / `setRewardTokens` | strat-behaviour-misc | Hardhat runs these across *all* strategies — needs a call: one representative test vs. a shared Foundry behaviour harness | |
Collaborator
There was a problem hiding this comment.
I don’t think these setters should be retested across every strategy. They are inherited from the shared InitializableAbstractStrategy implementation, so duplicating the same behavior suite for each strategy would add maintenance cost without meaningful additional confidence.
A single focused unit suite against one active representative strategy is sufficient to cover authorization, emitted events, state updates, and invalid reward-token addresses. Strategy-specific overrides or disabled functions can remain covered in their respective strategy suites
clement-ux
reviewed
Jul 21, 2026
|
|
||
| | Item | Area | Status | | ||
| |------|------|--------| | ||
| | Beacon exact SSZ roots + `0x01` validator vector | beacon | **Deferred** — byte-exact proof fixtures are substantially more complex and will be handled separately. | |
Collaborator
There was a problem hiding this comment.
Update: this deferred item is now implemented in Foundry.
Coverage added:
test_merkleizeSignature_knownValueuses the exact BLS signature vector and asserts the SSZ root0x5b449f....test_merkleizePendingDeposit_knownValuederives the Beacon-format public-key hash from the real vector and asserts the exact pending-deposit root0xc27ca5....test_verifyValidator_0x01WithdrawalCredentialsvalidates Hoodi validator1222119, including its0x01withdrawal credentials and full 1,696-byte / 53-witness Merkle proof.
These are deterministic unit tests and do not require a Hoodi fork or Beacon RPC. The targeted BeaconProofsLib suite passes all 50 tests, and the complete Foundry unit suite passes all 2,124 tests.
Related commits:
clement-ux
reviewed
Jul 21, 2026
|
|
||
| ## Batch 1 — implemented (2026-07-20) | ||
|
|
||
| 18 tests added against existing Foundry infrastructure on configured fork chains, all passing (full |
Collaborator
There was a problem hiding this comment.
Batch 1 added 18 focused tests using the existing Foundry fixtures:\n\n- 2 WOETH zero-amount cases for redeem(0) and withdraw(0).\n- 5 CompoundingStakingSSV initial-deposit configuration cases: default value, successful update/event, unauthorized caller, below-minimum and above-maximum values.\n- 1 exact OUSD trustee-fee regression test for the 1e9 fee case that the previous tolerance could not detect.\n- 4 Merkl pool-booster authorization tests for the campaign type, reward token, distributor and campaign-data setters.\n- 6 mainnet smoke assertions across the OUSD/OETH Curve AMOs and MorphoV2: Timelock governor, CRV reward token and configured harvester invariants.\n\nImplementation and documentation commits:\n- f6baa361d — adds all 18 Batch 1 tests.\n- 52837bfc8 — records the agreed skip/defer decisions.
clement-ux
reviewed
Jul 21, 2026
|
|
||
| ## Batch 2 — implemented (2026-07-21) | ||
|
|
||
| 20 focused tests added, with the exact Beacon SSZ vectors intentionally left for a later batch: |
Collaborator
There was a problem hiding this comment.
Batch 2 added 20 focused tests:\n\n- 4 Curve AMO fork scenarios covering attacker-added WETH/OETH and USDC/OUSD liquidity, with protocol profit measured as delta totalValue - delta totalSupply and explicit USDC decimal scaling.\n- 4 OUSD fuzz properties covering transfers and mints involving yield-delegation sources and targets while preserving accounting and supply invariants.\n- 9 shared strategy-configuration unit tests for setHarvesterAddress and setRewardTokenAddresses, including governor/strategist permissions, events, state changes and invalid inputs.\n- 3 Base/HyperEVM TimelockController fork tests covering schedule/execute behavior, idempotence and previously scheduled operations.\n\nImplementation and documentation commits:\n- 618ef900c — OUSD delegation fuzzing and representative strategy configuration.\n- 874925f1f — strengthened OETH and OUSD Curve AMO front-running coverage.\n- 54b3103db — Base and HyperEVM TimelockController support and fork coverage.\n- 2d8d65589 — records the Batch 2 implementation status in this report.
Collaborator
|
Added Foundry coverage for the three 21-validator Compounding SSV scenarios using the captured Beacon proofs from compoundingSSVStaking-validatorsData.json. The fixture deploys the real EnhancedBeaconProofs implementation, prepares all 21 validators at the historical strategy address, and verifies exact validator balances and pending deposits from snapshot index 5. It covers no pending deposits, two deposits on a zero-balance validator, and mixed deposits across active and zero-balance validators. Verification:
|
Closed by 200e9cd (TwentyOneValidators.t.sol): the three multi-validator real-Beacon-proof scenarios now covered. strat-compounding-ssv 75->78 covered, 26->23 missing; Total 1071->1074 / 434->431. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
clement-ux
added a commit
that referenced
this pull request
Jul 23, 2026
* Add Foundry build system with Soldeer dependency management
Set up forge compilation for all 302 Solidity files alongside existing
Hardhat config. Add foundry.toml with Soldeer deps (forge-std, OZ 4.4.2,
Chainlink CCIP, LayerZero v2, solidity-bytes-utils) and transitive
remappings for LZ dependency chain. Replace hardhat/console.sol import
with forge-std/console2.sol in MockRebornMinter.
* Add mock contract declarations to Base test contract
Add MockStrategy and MockNonRebasing state variables and imports
to the shared Base contract so they are available to all test suites.
* Add OUSDVault shared test setup and mint tests
Shared.sol deploys OUSD + OUSDVault behind proxies, configures the
vault with withdrawal delay, rebase rate max, and drip duration,
then funds matt and josh with 100 OUSD each.
Mint.t.sol covers mint, deprecated mint overload, mintForStrategy,
burnForStrategy, and auto-allocate on mint (13 tests).
* Add OUSDVault governance tests
Cover governor(), isGovernor(), two-step transferGovernance +
claimGovernance flow, and access control reverts (9 tests).
* Add OUSDVault view function tests
Cover totalValue, checkBalance, getAssetCount, getAllAssets,
getStrategyCount, getAllStrategies, isSupportedAsset, and the
deprecated oUSD() accessor (18 tests).
* Add OUSDVault admin tests
Cover all VaultAdmin setters, pause functions, strategy management,
and token rescue. Includes revert paths for unauthorized callers,
invalid values, and edge cases like "Asset not supported by Strategy",
"Strategy has funds", and "Parameter length mismatch" (98 tests).
* Add OUSDVault rebase and yield tests
Cover rebase pausing, yield distribution to rebasing accounts,
non-rebasing exclusion, trustee fee accrual, previewYield,
drip duration smoothing, _nextYield early-return branches, and
the defensive fee >= yield check (19 tests).
* Add OUSDVault allocation tests
Cover allocate to default strategy, vault buffer, withdrawal queue
reserves, depositToStrategy, withdrawFromStrategy,
withdrawAllFromStrategy, withdrawAllFromStrategies, capital-paused
revert, and early return when no asset available (23 tests).
* Add OUSDVault withdrawal queue tests
Cover requestWithdrawal, claimWithdrawal, claimWithdrawals,
addWithdrawalQueueLiquidity, strategy-queue interactions,
mint-covers-outstanding scenarios, full drain edge cases,
insolvency and slash scenarios, solvency at 3%/10%
maxSupplyDiff, rebase-on-redeem, and capital-paused claims
(55 tests).
* Configure Foundry test infrastructure
Add test directory, solmate dependency, and remappings to
foundry.toml. Add .gitkeep placeholders for the test directory
structure. Add lcov.info to .gitignore.
* Move test helpers to bottom of files and remove unused import
* Extract npm tgz dependency installation to shell script
* Add unit-test skill for generating Foundry tests
Captures the established test conventions (directory layout, inheritance
chain, naming patterns, helper idioms, fuzz config) so future contract
test suites follow the same structure as OUSDVault.
* Add OUSDVault fuzz tests and Foundry fuzz configuration
Add fuzz tests for mint, rebase, and withdraw covering key properties
(scaling, round-trip recovery, yield distribution, queue invariants).
Configure foundry.toml with 1024 runs and deterministic seed.
* docs(skill): add commit automation skill
* test(vault): add OETHVault Foundry unit tests with full branch coverage
* test(token): add OUSD Foundry unit tests and improve branch coverage to 96%
- Add Burn.t.sol and Initialize.t.sol for previously untested functions
- Add missing revert tests to Mint, TransferFrom, Transfer, YieldDelegation
- Fix truncated assertion in Mint.t.sol
- Update unit-test skill to enforce one file per function rule
* docs(skill): add coverage requirements to unit-test skill
Enforce minimum coverage thresholds: 100% functions, 90% branches/lines/statements.
Includes iterative improvement workflow and requires explanation for uncovered paths.
* test(token): add WOETH Foundry unit tests with 100% function coverage
* fix(skill): ensure commit skill stages untracked files
* test(token): add OETH, OETHBase, and OSonic view function tests
* test(token): add wrapped token unit tests (WOETHBase, WOETHPlume, WOSonic, WrappedOusd)
Add Foundry unit tests for remaining wrapped token variants and declare
their state variables in Base.sol for shared test infrastructure.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(zapper): add Foundry unit tests with 100% coverage for all zapper contracts
Add comprehensive unit tests for OETHZapper, OETHBaseZapper, OSonicZapper,
and WOETHCCIPZapper. Tests cover all public functions, revert conditions,
event emissions, and edge cases. Uses vm.mockCall for CCIP router and
vm.etch for hardcoded addresses (wS, Base WETH).
35 tests across 9 test suites — 100% line/statement/branch/function coverage.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(poolBooster): add Curve pool booster Foundry unit tests with 100% coverage
Add 120 tests across 22 files covering CurvePoolBooster, CurvePoolBoosterPlain,
and CurvePoolBoosterFactory. Includes concrete tests for all public/external
functions and fuzz tests for fee handling and salt encoding.
Also adds shared test infrastructure: Base.sol pool booster state vars,
MockCreateX for deterministic CREATE2 testing, and naming convention rules.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(poolBooster): add Merkl pool booster Foundry unit tests
Add unit tests for PoolBoosterMerkl and PoolBoosterFactoryMerkl covering
constructor, bribe, isValidSignature, getNextPeriodStartTime, factory
create/compute, and setMerklDistributor. Includes fuzz test for period timing.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(poolBooster): add SwapX pool booster Foundry unit tests
Add unit tests for PoolBoosterSwapxSingle, PoolBoosterSwapxDouble, and their
factories. Covers constructor validation, bribe mechanics, split calculations,
factory create/compute, and abstract factory functions (bribeAll, removePoolBooster).
Includes fuzz test for double bribe split amounts.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(poolBooster): add Metropolis pool booster Foundry unit tests
Add unit tests for PoolBoosterMetropolis and PoolBoosterFactoryMetropolis
covering constructor, bribe mechanics with rewarder/voter mocking, and
factory create/compute functions.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(proxies): add Foundry unit tests for proxy contracts with 100% coverage
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(tests): rename Base.sol and Shared.sol to .t.sol extension
Ensures forge excludes test helper files from coverage reports.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(governance): add Foundry unit tests for governance contracts with 100% coverage
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(skill): ensure commit skill always runs git commit before asking questions
* test(strategies): add Foundry unit tests for VaultValueChecker, BridgedWOETH, and Generalized4626 strategies
- 22 new test files with 109 tests across concrete, fuzz, and view function categories
- Deploy real OUSD/OUSDVault and OETH/OETHVault instead of mocks wherever possible
- VaultValueChecker: 100% coverage (lines, statements, branches, functions)
- BridgedWOETHStrategy: 100% lines/statements/functions, 87% branches (3 unreachable require(false))
- Generalized4626Strategy: 100% coverage (lines, statements, branches, functions)
- Move shared state variables (ousdChecker, oethChecker, bridgedWOETHStrategy) to Base.t.sol
* test(strategies): add Foundry unit tests for BaseCurveAMOStrategy and CurveAMOStrategy
- Add comprehensive concrete + fuzz tests for both Curve AMO strategies
- Create mock contracts: MockCurvePool, MockCurveGauge, MockCurveGaugeFactory, MockCurveMinter
- Add BaseCurveAMOStrategy and CurveAMOStrategy state variables to Base.t.sol
- Update SKILL.md to prohibit --ir-minimum on forge coverage
* test(strategies): add Foundry unit tests for SonicStakingStrategy and SonicSwapXAMOStrategy
Add comprehensive unit tests (concrete + fuzz) for Sonic strategy contracts
with mocks for WrappedSonic, SwapXPair, SwapXGauge, and enhanced MockSFC.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* docs(skill): update unit-test skill with product vault types, mock conventions, and coverage guidelines
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(strategies): add Foundry unit tests for CrossChainMasterStrategy and CrossChainRemoteStrategy
- 217 tests covering deposit, withdraw, relay, CCTP message processing, admin, and view functions
- Includes concrete tests, fuzz tests, constructor validations, and relay security checks
- 100% line/statement/function coverage, 90%+ branch coverage
- Added MockFailableERC4626Vault for testing deposit/withdraw failure catch blocks
- Updated unit-test skill to check for existing Hardhat tests first
* test(strategies): add Foundry unit tests for AerodromeAMOStrategy
- 81 tests covering deposit, depositAll, withdraw, withdrawAll,
rebalance, collectRewardTokens, admin, view functions, and
constructor validation
- Mocks: MockCLPool, MockNonfungiblePositionManager, MockCLGauge,
MockSwapRouter, MockSugarHelper
- Uses real OETHBase + OETHBaseVault deployed via proxies
- Covers paths absent from Hardhat fork tests: constructor requires,
OETHb-swap vault-minting path, NotEnoughWethLiquidity,
Unexpected token owner, Non zero wethAmount, gauge restake
guard when liquidity is zero after full withdrawal
* chore: add test infrastructure for NativeStaking strategies
Add imports, state variables, and mocks needed by NativeStakingSSVStrategy
and CompoundingStakingSSVStrategy Foundry unit tests:
- Base.t.sol: declare MockSSVNetwork, MockSSV, MockDepositContract,
MockBeaconProofs, NativeStakingSSVStrategy, FeeAccumulator,
CompoundingStakingSSVStrategy, CompoundingStakingStrategyView
- MockSSVNetwork: add missing `withdraw` function
- MockBeaconRoots / MockWithdrawalRequest: precompile mocks for EIP-4788
and EIP-7002
- foundry.toml: allow fs read for JSON test data
- SKILL.md: refine coverage skip guidance
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(strategies): add Foundry unit tests for NativeStakingSSVStrategy
131 tests covering NativeStakingSSVStrategy, FeeAccumulator, and
ValidatorAccountant: accounting (30 concrete + 3 fuzz), manuallyFixAccounting
(18 concrete + 3 fuzz), reward collection (7), validator registration (4),
validator staking (5), validator exit (5), deposits (5 concrete + 2 fuzz),
withdrawals (7), configuration (13), SSV operations (5), check balance (3),
and receive ETH (2).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(strategies): add Foundry unit tests for CompoundingStakingSSVStrategy
125 tests covering CompoundingStakingSSVStrategy and CompoundingValidatorManager:
validator registration (12), validator staking (10), validator exit (10),
strategy balances (28), verify deposit (9), front-run/invalid detection (6),
slashed validator deposits (4), configuration (16), deposits (8 concrete +
2 fuzz), withdrawals (13), disabled functions (3), check balance (3), and
receive ETH (1). Validator data loaded from JSON at runtime via stdJson.
Includes README documenting remaining Hardhat test scenarios that require
real beacon proof data and should be ported as fork tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(automation): add Foundry unit tests for Safe automation modules
Full unit tests (100% function/branch coverage) for:
- AbstractSafeModule (via concrete harness)
- AutoWithdrawalModule (incl. fuzz tests)
- ClaimStrategyRewardsSafeModule
- CollectXOGNRewardsModule
- CurvePoolBoosterBribesModule
- ClaimBribesSafeModule
Light tests (constructor + access control) for:
- EthereumBridgeHelperModule
- BaseBridgeHelperModule
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(beacon): add Foundry unit tests for Endian, Merkle, and BeaconProofsLib
Add concrete and fuzz tests for three beacon chain library contracts:
- Endian: 100% coverage on byte-order conversion functions
- Merkle: 93-100% coverage on SHA256 merkle tree verification
- BeaconProofsLib: 98% line/statement coverage, 100% function coverage
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(strategies): add Foundry fork tests for CurveAMOStrategy
- Deploy fresh OETH/OETHVault/CurvePool/Gauge/Strategy on mainnet fork
- 64 fork tests across Deposit, Withdraw, Rebalance, CollectRewards
- Test real Curve pool math: deposit/withdraw with tilted pools,
improvePoolBalance modifier, solvency checks, heavily unbalanced pools
- Add fork-test skill with guidelines on fork vs unit test scope
- Add BaseFork helper, Addresses library, and ICurveStableSwapFactoryNG
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(strategies): add Foundry fork tests for AerodromeAMOStrategy
37 fork tests covering deposit, withdraw, rebalance, and reward collection
against real Aerodrome Slipstream pools on Base. Includes quoter-driven
price manipulation tests and all custom error revert paths.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(strategies): add Foundry fork tests for CrossChain strategies
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(strategies): add Foundry fork tests for SonicStakingStrategy
Migrate SonicStakingStrategy fork tests from Hardhat to Foundry with 18
fork-worthy tests covering initial state, deposit, rewards, undelegation,
SFC withdrawal (including slashing scenarios), withdraw, and checkBalance.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(strategies): add Foundry fork tests for SonicSwapXAMOStrategy
Migrate 1635 lines of Hardhat JS fork tests to 72 Foundry tests covering
deposit, withdraw, rebalance, reward collection, front-running, and
initial state scenarios. Deploys fresh OSonic, OSVault, SwapX pool (via
factory createPair), gauge (via Voter createGauge), and strategy — no
bytecode cloning or storage hacks needed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* docs(skill): add codex repo skills
* test(strategies): add Foundry fork tests for NativeStakingSSVStrategy
Migrate NativeStakingSSVStrategy fork tests from Hardhat to Foundry,
targeting Strategy 2 (operators [752, 753, 754, 755]). Covers deposit,
validator registration/exit, accounting, and harvest flows.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test: add foundry beacon proofs fork tests
* test: add foundry beacon roots fork tests
* test: add foundry partial withdrawal fork tests
* test(poolBooster): add Foundry fork tests for CurvePoolBooster
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test(poolBooster): add Foundry fork tests for SwapX, Merkl, Metropolis and Shadow pool boosters
Migrate pool booster fork tests from Hardhat to Foundry:
- SwapX (Double/Single): bribe, bribeAll, create, remove, address computation
- Shadow: bribe via Shadow gauge with NotifyReward event
- Merkl (Mainnet): create, bribe twice, skip below min amounts, deployment params
- Metropolis: create, bribe twice, skip below min amounts
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* test(automation): add Foundry fork tests for Safe modules
Migrate EthereumBridgeHelperModule, BaseBridgeHelperModule, and
ClaimStrategyRewardsSafeModule Hardhat fork tests to Foundry.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(test): pin NativeStakingSSV fork tests to block 24640000
The strategy was upgraded after this block, removing
registerSsvValidators. Add a block-pinned overload to BaseFork.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Revert "fix(test): pin NativeStakingSSV fork tests to block 24640000"
This reverts commit 2189903fa72d6fdec3b91197501f07e48d5ff486.
* test(NativeStakingSSV): skip fork tests pending SSV ETH-payment migration
The on-chain strategy was upgraded to match SSV Network's migration from
SSV-token to ETH-based payments, causing selector mismatches with our
source code. Skip tests until the contract interfaces are updated.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* docs(skill): add smoke-test skill for Claude Code and Codex
* feat(deploy): add DeployManager and Resolver infrastructure
Foundry-based deployment pipeline with JSON-driven script execution,
deterministic Resolver for contract address resolution, and governance
simulation support for smoke testing.
* test(smoke): add OUSD smoke tests with DeployManager bootstrapping
Verifies deployment health for OUSD by resolving contracts from the
Resolver and testing mint, redeem, transfer, rebasing, yield delegation,
and view functions against real mainnet fork state.
* feat(deploy): add Base chain support to DeployManager
* test(smoke): add smoke tests for OETH, OETHBase, and wrapped tokens
- OETH: ViewFunctions, VaultViewFunctions, Mint, Redeem, Transfer,
Rebasing, YieldDelegation
- OETHBase: same feature coverage
- wrappedToken: ViewFunctions, SharePrice, DepositRedeem for WOETH,
WOSonic, and WrappedOUSD
- OUSD: VaultViewFunctions (vault-level view checks)
* test(smoke): add OSonic smoke tests, skipped pending vault upgrade
The on-chain OSVault was deployed before the vault refactoring that
introduced mint(uint256), asset(), and oToken(). Tests are written
against the target interface but skipped via vm.skip(true) until a
Sonic upgrade deploy script brings the vault up to date.
* test(smoke): extend OUSD smoke tests with additional scenarios
- Rebasing: add opt-in/opt-out loop inflation check and governance
rebaseOptIn test
- Transfer: add full-balance and self-transfer tests
- YieldDelegation: add source-can-transfer and undelegate-preserves-
yield tests
* fix(smoke): use additive deal in _ensureVaultLiquidity
The conditional deal could silently skip funding when the vault already
held enough token balance — but that balance was fully allocated to prior
claimable withdrawal requests, leaving nothing for the new request and
causing claimWithdrawal to revert with "Queue pending liquidity".
Switch to an additive deal (currentBalance + needed) so the shortfall is
always covered on top of whatever is already in the vault. Applied to
OUSD, OETH, and OETHBase shared helpers (OSonic was authored correctly
from the start).
* feat(deploy): add example mainnet deployment script for OUSD upgrade
Demonstrates the three-phase lifecycle (execute → governance proposal →
fork verification) as a template for future mainnet deployments.
Also tracks deployment registry JSON files and refines .gitignore to
include deployments-*.json while excluding fork variants.
* test(smoke): add WOETHBase wrapped token smoke tests
* chore: remove Echidna fuzzing contracts
These property-based fuzzing contracts were failing when run under
Forge's test runner (constructor reverts with "Caller is not the
Governor") and are not part of the Foundry test suite.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* chore(smoke): reorganize wrapped token tests and fix WOSonic roundtrip tolerance
Move wrapped token smoke tests from tests/smoke/wrappedToken/ to
tests/smoke/token/ to consolidate all token smoke tests under a single
directory, matching the OSonic layout.
Also fix test_convertToShares_roundtrip tolerance from 1 to 2 wei in
WOSonic to account for ERC4626 rounding in the convertToShares →
convertToAssets roundtrip.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(deploy): compile scripts/ directory and fix 000_Example template
- Add `script = "scripts"` to foundry.toml so Forge compiles all
deploy scripts into out/, which DeployManager requires to load them
at runtime via vm.deployCode()
- Add missing `using GovHelper for GovProposal` in AbstractDeployScript
(documented in ARCHITECTURE.md but absent from the code)
- Fix 000_Example.s.sol: add GovHelper/GovProposal imports, add
`using GovHelper for GovProposal`, and add missing payable cast on
proxy address
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(deploy/sonic): add 026_VaultUpgrade deploy script
Migrates the Hardhat deploy/sonic/026_vault_upgrade.js to Foundry.
_execute() deploys a new OSVault implementation with wS as the base
asset. _buildGovernanceProposal() is intentionally empty because Sonic
uses a Timelock Controller rather than the mainnet Governor. _fork()
pranks the Sonic timelock to upgrade the vault proxy and set
SonicStakingStrategy as the default strategy, then asserts the
implementation was updated.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* test(smoke): enable OSonic smoke tests
The OSonic smoke tests were blocked behind vm.skip(true) pending the
026_VaultUpgrade deploy script, which upgrades the on-chain OSVault to
an implementation that exposes mint(uint256), asset(), and oToken().
- Remove vm.skip(true) now that 026_VaultUpgrade.s.sol is in place
- Fix _ensureVaultLiquidity to use additive deal (add on top of existing
balance rather than replace it), matching the OETH/OETHBase pattern.
The existing vault balance may already be fully allocated to prior
claimable requests and must not be overwritten.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* test(smoke): add CurvePoolBoosterFactory and CurvePoolBoosterPlain smoke tests
- Add CurvePoolBoosterFactory and CurvePoolBoosterPlainOETH address
constants to Mainnet library in Addresses.sol
- Add shared setup forking mainnet and instantiating live contracts
- Add 11 factory tests: view functions (governor, strategist, registry,
pool booster list/mapping, address computation, salt encoding) and
mutative ops (create booster, remove booster)
- Add 13 plain booster tests: view functions (rewardToken, gauge, fee,
campaignRemoteManager, votemarket, targetChainId) and mutative ops
(createCampaign, manageCampaign, closeCampaign)
* chore(deploy): register automation and pool booster contracts in deployment JSONs
Add contract addresses for automation safe modules (Base and mainnet),
BRIDGED_WOETH, BRIDGED_WOETH_STRATEGY_PROXY, and pool booster contracts
to the deployment registry so smoke tests can resolve them via the
DeployManager resolver instead of hardcoded addresses.
Remove CurvePoolBoosterFactory and CurvePoolBoosterPlainOETH from
Addresses.sol since they now live in the deployment registry.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* test(smoke): add automation safe module smoke tests
Add smoke tests for BaseBridgeHelperModule, EthereumBridgeHelperModule,
and ClaimBribesSafeModule covering both view functions and mutative flows.
BaseBridgeHelperModule (14 tests): depositWOETH, depositWOETHAndClaimWithdrawal,
depositWETHAndRedeemWOETH, bridgeWETHToEthereum, bridgeWOETHToEthereum,
depositWETHAndBridgeWOETH, claimAndBridgeWETH.
EthereumBridgeHelperModule (11 tests): mintAndWrap, bridgeWOETHToBase,
bridgeWETHToBase, mintWrapAndBridgeToBase. Three functions not yet in the
deployed bytecode (unwrapAndRequestWithdrawal, claimWithdrawal,
claimAndBridgeToBase) are omitted until the contract is redeployed.
ClaimBribesSafeModule: tests skip gracefully if the contract is not yet
deployed at the registered address.
All contracts resolved via DeployManager resolver, not hardcoded addresses.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(smoke/poolBooster): resolve contracts via DeployManager resolver
Replace hardcoded Mainnet.CurvePoolBoosterFactory and
Mainnet.CurvePoolBoosterPlainOETH addresses with resolver.resolve() calls,
consistent with the pattern used across all smoke tests.
Also clean up the unused CrossChain import and the aliased import of
CurvePoolBoosterFactory, and replace Mainnet.CurvePoolBoosterPlainOETH
references in assertions with address(curvePoolBoosterPlain).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* test(smoke): add pool booster smoke tests for SwapxSingle, SwapxDouble, Metropolis, Merkl, and CentralRegistry
Add smoke tests across Sonic, Mainnet, and Base for all non-Curve pool booster
contracts. Register factory, registry, and representative booster addresses in
deployment JSONs for each chain.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* test(smoke): add AerodromeAMOStrategy smoke tests
Add smoke test coverage for the AerodromeAMOStrategy on Base chain,
verifying that the live deployment (strategy + pool + gauge) is healthy.
- ViewFunctions (23 tests): immutables, configuration, position state, gauge staking
- Deposit (6 tests): checkBalance increases, auto-rebalance when in range, access control
- Rebalance (8 tests): no-swap/quoter-based rebalance, LP restaked, no residual tokens
- Withdraw (8 tests): partial withdraw, withdrawAll, full lifecycle, access control
- CollectRewards (2 tests): harvester can collect, access control
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* test(smoke): add CrossChainMasterStrategy smoke tests
Validate deployment health of CrossChainMasterStrategy via DeployManager/Resolver
with view function checks, deposit, withdraw, balance check, and token received tests.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* test(smoke): add CrossChainRemoteStrategy smoke tests
Validate deployment health of the remote strategy on Base chain (counterpart
to the existing CrossChainMasterStrategy smoke tests on mainnet). Covers view
functions, deposit/withdraw relay flows, balance updates, and relay validation
guards.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* test(smoke): add SonicStakingStrategy smoke tests
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* docs(skill): enforce Resolver-only contract resolution in smoke-test skill
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* test(smoke): add CurveAMO strategy smoke tests for OETH, OUSD, and Base
Add smoke test suites for three CurveAMO strategy deployments:
- OETH Curve AMO (mainnet) — CurveAMOStrategy at 0xba0e352
- OUSD Curve AMO (mainnet) — CurveAMOStrategy at 0x26a02ec
- superOETHb Curve AMO (Base) — BaseCurveAMOStrategy at 0x9cfcAF8
Each suite covers ViewFunctions, Deposit, Withdraw, Rebalance
(mintAndAddOTokens, removeAndBurnOTokens, removeOnlyAssets), and
CollectRewards. Rebalance tests use dynamic pool tilting helpers
that read current pool state to ensure correct preconditions.
Also adds resolver entries to deployments-1.json and deployments-8453.json.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* test(smoke): add SonicSwapXAMOStrategy smoke tests and remove RevertWhen tests
Add smoke tests for SonicSwapXAMOStrategy (Deposit, Withdraw, Rebalance,
CollectRewards, ViewFunctions) and register the strategy proxy in
deployments-146.json. Remove all RevertWhen tests from AerodromeAMO and
SwapXAMO smoke suites since access control and input validation belong
in unit tests, not smoke tests. Update the smoke-test skill to enforce
this rule.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(js): bypass Lodestar client for beacon state SSZ fetch
The Lodestar API client (v1.38.0) Accept header allows JSON fallback,
causing ERR_ENCODING_INVALID_DATA when the beacon node returns binary
SSZ data with a JSON content-type. Use direct HTTP fetch with SSZ-only
Accept header and ArrayBuffer reading to avoid text decoding issues.
* ci(foundry): add Foundry CI workflow with 10 jobs
Add a standalone Foundry CI pipeline (fmt, build, unit-tests, coverage,
fork-tests and smoke-tests per chain). Fork/smoke jobs dynamically
discover test directories by grepping for chain-specific fork functions,
requiring zero maintenance as new tests are added. Scheduled daily cron
runs fork/smoke tests to catch on-chain state changes.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(foundry): update tests for master merge contract changes
- PoolBoosterMerkl → PoolBoosterMerklV2 (BeaconProxy pattern with initialize)
- SonicSwapXAMOStrategy constructor reduced to 2 params (oToken/asset from vault)
- CompoundingStakingSSVStrategy constructor removed _ssvToken param
- registerSsvValidators/registerSsvValidator removed ssvAmount param
- depositSSV/withdrawSSV replaced by migrateClusterToETH
- manageBribes now requires selectedPoolBoosters array
- ISwapXPair/ISwapXGauge renamed to IAlgebraPair/IAlgebraGauge
- snapBalances/verifyBalances now onlyRegistrator
- CCTPMessageTransmitterMock2 constructor takes peerDomainId
- MockERC4626Vault/MockSSVNetwork extended for new interfaces
- Smoke tests use low-level calls for V1/V2 compatibility
* fix: remove forge-std import from MockRebornMinter breaking Hardhat CI
* fix(ci): use forge soldeer subcommand and guard beaconProofsFixture from mocha
- soldeer is now a forge subcommand, not a standalone binary
- beaconProofsFixture.js ran main() when loaded by mocha during coverage
* fix(ci): skip tests/ directory in foundry contract size check
* fix(ci): also skip scripts/ directory in foundry contract size check
* fix(ci): fix foundry CI failures for coverage, fork, and smoke tests
- Skip aerodrome strategy in coverage to avoid stack-too-deep errors
- Replace /bin/zsh with /bin/bash in BeaconRoots FFI calls for Ubuntu runners
- Add forge build to foundry-setup action so deploy script artifacts exist for smoke tests
- Add .gitkeep to scripts/deploy/base/ so the directory is tracked in git
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): fix remaining foundry CI failures
- Run forge fmt on all Solidity files to pass formatting check
- Fix coverage --skip to use regex substring ("strategies/aerodrome")
instead of invalid glob pattern ("*/strategies/aerodrome*")
- Add pnpm + Node.js setup to foundry-setup action so npm deps
(ethers) are available for BeaconProofs fork test FFI scripts
- Filter non-.s.sol files in DeployManager to prevent .gitkeep from
producing malformed artifact paths (out/.s.sol/$.json)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): revert forge fmt, keep targeted CI fixes only
Revert all forge fmt formatting changes. The formatting CI check will
remain failing — only keep targeted fixes:
- Coverage --skip regex pattern
- pnpm/Node.js setup for BeaconProofs FFI
- DeployManager .s.sol file filter for .gitkeep
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): replace .gitkeep with example deploy script for Base
Replace the .gitkeep file (which DeployManager would parse as a deploy
script producing a malformed artifact path) with a proper 000_Example
deploy script, matching the mainnet pattern. Revert the DeployManager
.s.sol filter as it's no longer needed.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): specify pnpm version 10 in foundry-setup action
pnpm/action-setup@v4 requires an explicit version since there is no
packageManager field in package.json. Match defi.yml which uses v10.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): add BEACON_PROVIDER_URL secret to fork-tests-mainnet job
The BeaconProofs fork test calls beaconProofsFixture.js via vm.ffi,
which uses @lodestar/api requiring BEACON_PROVIDER_URL. Without it
the Lodestar client fails with "Must set at least 1 URL in HttpClient".
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): remove coverage job, scope fmt to scripts/ and tests/ only
- Remove coverage CI job entirely (stack-too-deep issues)
- Scope forge fmt --check to only scripts/ and tests/ directories
- Format scripts/ and tests/ to pass the scoped check
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): format utils/beacon.js with Prettier
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): merge formatting/lint into single Foundry CI job
- Foundry "Formatting & Lint" job now runs:
- forge fmt --check on scripts/ and tests/ (Solidity)
- npx prettier on JS files (deploy, scripts, tasks, test, utils)
- pnpm run lint (eslint + solhint)
- Remove redundant "Contracts Linter" job from DeFi workflow
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(ci): add prettier check for Solidity in contracts/contracts/
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* chore(ci): remove Plume fork tests from DeFi workflow
* fix(ci): format EnhancedBeaconProofs.sol with Prettier
* test(supernova): add Foundry unit, fork, and smoke tests for OETHSupernovaAMOStrategy
- 18 unit test files (80 tests): constructor, deposit, withdraw, rebalance, fuzz
- 7 fork test files (72 tests): fresh pool/gauge via factory authorization, front-running, insolvency
- 6 smoke test files (19 tests): deployed contract validation via Resolver
- Infrastructure: add Supernova addresses to Addresses.sol, strategy to Base.t.sol, proxy to deployments-1.json
* feat(hyperevm): integrate HyperEVM into Foundry fork/smoke tests and deployment setup
- Add HyperEVM RPC endpoint to foundry.toml and fork helper in BaseFork.t.sol
- Add HyperEVM chain routing (ID 999) to DeployManager and Base.s.sol
- Create example deploy script in scripts/deploy/hyperevm/
- Create deployments-999.json with CrossChainRemoteStrategy address
- Add HyperEVM smoke tests (ViewFunctions, BalanceUpdate, Deposit, Withdraw, RelayValidation)
- Add fork-tests-hyperevm and smoke-tests-hyperevm CI jobs in foundry.yml
- Rename CrossChainRemoteStrategy smoke tests to CrossChainRemoteStrategyBase
* chore: remove unused UnitTests library from Addresses.sol
* chore: apply Prettier formatting to Solidity files
* chore(ci): add slither and snyk jobs to Foundry workflow
* perf(ci): speed up Foundry workflow with build caching and parallelism
- Cache forge build artifacts (out/, cache/) to avoid recompiling 1007 files each job
- Remove `needs: build` gate so test jobs start immediately in parallel
- Split Fork Tests (Mainnet) into 3 matrix chunks for parallel execution
- Skip foundry-setup entirely when no tests discovered for a chain
* test(morpho-v2): add Foundry unit, fork, and smoke tests for MorphoV2Strategy
- Unit tests (31): deposit, withdraw, withdrawAll override, maxWithdraw,
view functions, and fuzz tests with limited liquidity simulation
- Fork tests (20): fresh deploy against real Morpho V2 vault on mainnet
fork with mocked share guard for whitelist bypass
- Smoke tests (14): deployment health checks via DeployManager/Resolver
- Add MockMorphoV2Vault and MockMorphoV2Adapter for unit test isolation
- Register MORPHO_OUSD_V2_STRATEGY_PROXY in deployments-1.json
* perf(ci): commit BeaconProofs fixture to eliminate 10min beacon RPC calls
- Commit pre-generated beacon proof fixture for slot 12235962 (20KB JSON)
- Read fixture via vm.readFile() with FFI fallback for custom slots
- Narrow forge build cache to exclude 273MB beacon SSZ state files
- Reduce Mainnet fork chunks from 3 to 2 now that BeaconProofs is instant
* test(consolidation): add Foundry unit, fork, and smoke tests for ConsolidationController
- Fork tests extend BaseFork with fresh OETH+OETHVault deployment
- Strategy proxies upgraded to new impls pointing at fresh vault
- Unit tests cover request/fail/confirm consolidation and operations
- Smoke tests validate deployed configuration and forwarding
- Add missing addresses to Addresses.sol (FeeAccumulators, BeaconProofs)
* refactor(ci): reorganize Foundry tests into chain-specific directories and fix HyperEVM smoke tests
- Move fork/smoke tests into chain-specific subdirs (mainnet/, base/, sonic/, hyperevm/)
- Update CI workflow to use `find` with chain-specific paths instead of `grep -rl`
- Fix HyperEVM smoke tests: mock CCTP MessageTransmitter and TokenMessenger
to avoid arithmetic overflow from real on-chain CCTP contracts
- Run forge fmt on all test/script files to fix formatting
* chore: add Makefile and .env.example, clean up dev.env
- Add Makefile with targets for build, test, deploy, coverage, and gas
- Multi-chain support (mainnet, base, sonic, hyperevm) for test and deploy
- Reorganize dev.env with consistent sections and annotations
* chore(ci): remove legacy defi workflow
* fix(test): increase ERC-4626 roundtrip tolerance to 2 wei in smoke tests
* fix(test): wrap exitSsvValidator in try-catch for SSV state changes
The fork test was failing because the on-chain SSV validator state had
changed, causing IncorrectValidatorStateWithData to revert before
reaching the existing try-catch on removeSsvValidator. Both SSV calls
are now wrapped since the test validates access control, not SSV state.
* fix(test): handle SSV state changes in NoConsolidation fork tests
Wrap exitSsvValidator calls in try-catch to handle
IncorrectValidatorStateWithData from the SSV Network when validators
have already been exited on-chain. Falls back to vm.store where the
test needs the EXITING state for subsequent assertions.
* perf(test): slim down Base.t.sol and add test profile for faster compilation
Base.t.sol imported ~60 contracts and was inherited by ~99 Shared.t.sol
files. Any change to it invalidated the entire Foundry cache, triggering
a full recompile. Strip Base.t.sol to only actors, constants, IERC20
external tokens, fork IDs, and setUp(). Move typed contract/proxy/mock
variable declarations into each Shared.t.sol that actually uses them.
Also add a [profile.test] section to foundry.toml with optimizer disabled
and dynamic_test_linking enabled for faster test-only compilation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* chore: remove test profile from foundry.toml
* style(test): reformat profit calculation in FrontRunning tests
* test(smoke): add vault smoke tests for OUSD, OETH, OETHBase, and OSonic
Verify deployment health for all four vault contracts via DeployManager/Resolver:
- ViewFunctions: governor, strategist, default strategy, vault buffer, claim delay, pause states
- Mint: totalValue increase, asset debit, vault receives asset
- Rebase: succeeds, increases supply, previewYield correctness
- Allocate: depositToStrategy, withdrawFromStrategy, totalValue preservation
- WithdrawalQueue: metadata updates, multi-claim, liquidity management, request storage
* feat: add storage layout compatibility checker for proxy upgrades
Standalone Node.js script that compares storage layouts between git refs
using forge inspect. Reviewers can run it during PR review to verify
upgrades won't cause storage slot conflicts.
* fix: handle forge output quirks and gap-carving in storage checker
- Add --json flag to forge inspect and strip tracing logs from stdout
- Install soldeer dependencies before building in worktrees
- Make forge build non-fatal (partial builds can still inspect some contracts)
- Handle the "carve from gap" upgrade pattern: new variables replacing
the start of a __gap array are valid if the gap shrinks by the exact
number of slots used
* fix: use install-deps.sh, forge clean, and extra_output for storage checker
- Use install-deps.sh instead of forge soldeer install (handles npm tgz packages)
- Run forge clean before building to avoid stale cache issues
- Add extra_output_files = ["storageLayout"] to foundry.toml
* perf: use forge inspect --force instead of full forge build
Only compiles the target contract and its dependencies, not the entire
repo. Reduces runtime from ~3min to ~10s per contract check.
* refactor: DRY up Makefile with pattern rules, targeted builds, and usage examples
* style: format check-storage-layout.js with prettier
* fix(test): widen maxSupplyDiff in Base smoke tests to handle deal-induced backing drift
- _ensureVaultLiquidity now sets maxSupplyDiff to 10% so that
artificially dealt WETH (which inflates totalValue beyond what the
drip-limited rebase can match) doesn't trip the _postRedeem check.
- Add _ensureAssetAvailable helper to cover outstanding withdrawal
queue obligations before depositToStrategy calls.
- Fixes Smoke_Concrete_OETHBase_Redeem_Test and unblocks the previously
hidden Smoke_Concrete_OETHBaseVault_Allocate_Test failures.
* Foundry migration improve speed (#2868)
* refactor(test): use interfaces instead of concrete imports in OUSDVault tests
Replace concrete contract imports (OUSDVault, OUSD, VaultStorage, Proxies)
with interface-only imports (IVault, IOToken, IProxy) and vm.deployCode.
This keeps test compilation units small for better Forge caching.
- Add IOToken and IProxy interfaces
- Add isGovernor() to IVault interface
- Update Shared.t.sol to deploy via vm.deployCode and cast to interfaces
- Replace all VaultStorage.Event with IVault.Event references
- Use struct field access instead of tuple destructuring
- Document interface-only testing pattern in tests/README.md
* refactor(test): use interfaces instead of concrete imports in OETHVault tests
Same migration as OUSDVault: replace concrete contract imports
(OETHVault, OETH, VaultStorage, Proxies) with interface-only imports
(IVault, IOToken, IProxy) and vm.deployCode for better Forge caching.
- Update Shared.t.sol to deploy via vm.deployCode and cast to interfaces
- Replace all VaultStorage.Event with IVault.Event references
- Use struct field access instead of tuple destructuring
* refactor(test): use interfaces instead of concrete imports in OUSD token tests
Replace concrete contract imports (OUSD, OUSDVault, Proxies) with
interface-only imports (IOToken, IVault, IProxy) and vm.deployCode
for better Forge caching.
- Update Shared.t.sol to deploy via vm.deployCode and cast to interfaces
- Replace all OUSD.Event with IOToken.Event references
- Update Initialize.t.sol to use vm.deployCode for fresh deployments
- Update Transfer.t.sol MockNonRebasingTwo helper to use IOToken
* refactor(test): use interfaces instead of concrete imports in OETH/OETHBase/OSonic token tests
Replace concrete token imports with IOToken interface and vm.deployCode
for better Forge caching.
* refactor(test): use interfaces instead of concrete imports in wrapped token tests
Add IWOToken interface for WOETH/WOETHBase/WOETHPlume/WOSonic/WrappedOusd.
Replace concrete contract imports with interface-only imports (IWOToken,
IOToken, IVault, IProxy) and vm.deployCode across all wrapped token tests.
* docs: update test docs and unit-test skill for interface-only testing
- Fix vm.deployCode path typo in README
- Add proxy, token, and wrapped token deployment examples
- Add available interfaces reference table
- Update unit-test skill: interface types, vm.deployCode, checklist
* refactor(test): use interfaces instead of concrete imports in all strategy tests
Add 15 per-strategy interfaces in contracts/interfaces/strategies/ and
migrate all 15 strategy test suites to interface-only imports with
vm.deployCode for better Forge caching.
- ICurveAMOStrategy, IBaseCurveAMOStrategy, IOETHSupernovaAMOStrategy
- IAerodromeAMOStrategy, ISonicSwapXAMOStrategy, ISonicStakingStrategy
- IBridgedWOETHStrategy, IGeneralized4626Strategy, IMorphoV2Strategy
- INativeStakingSSVStrategy, ICompoundingStakingSSVStrategy
- IConsolidationController, ICrossChainMaster/RemoteStrategy
- IVaultValueChecker
* docs(skill): sync codex unit-test skill with interface-only testing rules
* refactor(test): migrate proxy unit tests to iProxy
* test: migrate poolbooster unit tests to interfaces
* test: migrate automation unit tests to interfaces
* prettier
* docs(skill): update fork-test skills with interface-only testing rules
* test(zapper): migrate zapper unit tests to interfaces
* test(origin): migrate base automation fork tests to interfaces
* test(base): migrate aerodrome amo fork tests to interfaces
* test(origin): finish base strategy fork migrations
* test(origin): migrate mainnet automation fork tests to interfaces
* test(origin): migrate pool booster fork tests to interfaces
* test(origin): migrate mainnet strategy fork suites
* test(sonic): migrate sonic fork suites
* docs(skill): update smoke-test skills with interface-only testing rules
* test(origin): migrate mainnet vault smoke tests
* test(origin): migrate smoke suites and unify staking interfaces
* test(origin): migrate smoke suites to interfaces
* test(origin): use make targets in foundry workflow
* chore(zapper): format zapper interfaces
* chore(origin): format strategy interface imports
* test(origin): fix abstract safe module constructor setup
* chore(origin): simplify forge test make targets
* feat(skill): add organize-test skill and apply to OUSDVault tests
Add a new Claude/Codex skill for reorganizing Foundry test files
structurally without semantic changes. Apply it to OUSDVault unit tests:
- Sort imports into named groups (Test base, External libraries, Project imports)
- Reorder state variables (CONSTANTS before CONTRACTS & MOCKS)
- Consolidate revert tests next to their parent function sections
* ci(origin): add unused import lint check to CI
Add `make lint-imports` step to the Formatting & Lint job in the Foundry
workflow so PRs introducing unused Solidity imports fail CI.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* chore(origin): remove 88 unused Solidity imports
Clean up unused imports across scripts, unit tests, fork tests, smoke
tests, and mocks flagged by `forge lint --only-lint unused-import`.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fmt
* chore: upgrade forge-std from 1.9.7 to 1.15.0
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* chore(test): remove NativeStakingSSVStrategy and ConsolidationController test suites
* refactor(test): centralize Foundry artifact paths (#2874)
* docs(skill): require referencing artifact paths through tests/utils/Artifacts.sol
Update unit-test and fork-test skills (Claude + Codex) to instruct that
vm.deployCode calls must use sub-library references like Vaults.OUSD or
Proxies.IG_PROXY instead of inline "contracts/...sol:Name" strings, and
to add missing entries to tests/utils/Artifacts.sol when needed.
* refactor(test): centralize Foundry artifact paths in tests/utils/Artifacts.sol
Introduce a single Artifacts.sol containing per-category sub-libraries
(Tokens, Vaults, Proxies) to replace inline "contracts/...sol:Name"
strings passed to vm.deployCode. Migrate OUSDVault unit test Shared.t.sol
as the first consumer; other test directories will be migrated incrementally.
* refactor(test): centralize foundry artifact paths
* chore: remove artifacts migration plan
* docs(test): document split artifact libraries
* refactor(test): organize foundry test imports
* lint
* fix(test): update collectRewardTokens tests for onlyHarvesterOrStrategist modifier
Align test expectations with master's PR #2853 which allows strategist
to call collectRewardTokens.
* chore(test): simplify mainnet fork workflow
* chore: remove AGENTS.md from .gitignore
* bring Foundry migration up to date with master
* Update foundry.toml to include additional ignored error codes and warnings
* Refactor: Remove deprecated test files and associated shared utilities for WOETHPlume and WOSonic
- Deleted unit tests for WOETHPlume: Deposit, ViewFunctions, and Shared utilities.
- Deleted unit tests for WOSonic: Deposit, ViewFunctions, and Shared utilities.
- Removed unused artifacts and references related to OSonic and WOETH in Proxies, Tokens, Vaults, Strategies, and Zappers.
- Cleaned up PoolBoosters artifacts by removing obsolete factory references.
* Add target to update Foundry fork block numbers in .env
* Remove deprecated Oracle contracts and associated documentation
- Deleted OSonicOracleRouter.sol and OracleRouter.sol as they are no longer needed.
- Removed related README.md files and SVG diagrams for OETH and Oracle routers.
- Updated generate.sh script to exclude Oracle-related UML generation.
- Cleaned up PlantUML files by removing references to obsolete Oracle components.
* Add IMerklDistributor import and update conditions in smoke tests for PoolBoosterMerkl
* Refactor: Update ClaimBribesSafeModule tests to skip if module is unavailable and add isModuleAvailable flag in shared test
* Remove Sonic fork and smoke tests from Foundry workflow and Makefile
* test: add missing foundry migration coverage
* fix: stabilize foundry ci checks
* fix: reduce morpho fork rpc load
* style: format import statements and improve code readability in tests
* style: match ci forge formatting
* Sparrow dom/foundry migration/talos actions (#2943)
* migrate all of the talos actions to viem and away from hardhat
* changing in place ethers references to remove hardhat
* remove unneeded files
* refactor(tasks): make block.js hardhat-free for standalone action runtime
- getBlock/getDiffBlocks resolve the provider via tasks/lib/network for the
standalone action CLI, with an hre fallback for legacy hardhat dev tasks
- lazy-require @nomicfoundation/hardhat-network-helpers (mine) so importing
block.js no longer loads hardhat; only advanceBlocks (test/dev) needs it
- completes the Talos hardhat removal: block.js was the last module pulling
hardhat onto the standalone action runtime path (tasks/lib/network merged
via #2943)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* add test report (#2942)
* add test report
* update test report
* fix error message
* add the easy tests
* update decisions
* test: cover OUSD delegation accounting and strategy config
* test(origin): strengthen curve amo front-running coverage
* feat(deploy): support Base and HyperEVM timelocks
* docs: record foundry migration batch two
* chore: update tests and defender action bundles
* chore: ignore generated defender bundles
* chore: match CI Forge formatting
* chore: format auto withdrawal task
* docs: record skip decisions in foundry migration open questions
Mark #5 (Algebra/Hydrex AMO), #7 (RebalancerModule), #10 (legacy OUSD
migration-state), #11 (whale withdrawAllFromStrategies) and #13
(WOETH-upgrade/EigenLayer/EIP-7702) as skipped; keeps Batch 2's #9 result.
Open: #3, #4, #12.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(ssv): cover 21 validators with real beacon proofs
* docs(gap): mark 21-validator SSV real-proof scenarios as closed
Closed by 200e9cda4 (TwentyOneValidators.t.sol): the three multi-validator
real-Beacon-proof scenarios now covered. strat-compounding-ssv 75->78 covered,
26->23 missing; Total 1071->1074 / 434->431.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Clément <clemmoller@gmail.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(deploy): add fresh OUSD deployment
* fmt
* fix(ci): match stable forge formatting
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Shahul Hameed <10547529+shahthepro@users.noreply.github.com>
Co-authored-by: Domen Grabec <grabec@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add foundry test report