[meyer9]

Summary

A test marks itself mock-only by opening with vm.skip(livePrecompiles), but nothing says when that's allowed: the reasons live only in per-test comments. This PR documents the criteria on the livePrecompiles flag in test/lib/BaseTest.sol, and adds a row pointing to them in LIVE_PRECOMPILE_TESTING.md. A test is mock-only when it:

1. writes storage at a fixed precompile address, which forge rejects with vm.store: cannot use precompile ... as an argument;
2. asserts a mock storage slot that the Rust implementation doesn't have, such as MockB20Storage.initializedSlot(); or
3. needs a state that no real sequence of calls can reach and that only mock storage writes can build.

It also says to branch on livePrecompiles when a behavior can be checked in both worlds. A test waiting on a base/base change should be gated behind an environment flag, like POLICY_DISPATCH_FIX, instead of this guard. base/base's live-precompile CI now leaves mock-only skips out of its skipped count (base/base#6154), so misusing the guard would hide the test there.

How the current skips fit

I removed every vm.skip(livePrecompiles) and ran the tests against the live precompiles (base-anvil at 541626fd, Denim, base-std 7321d891):

Criterion	Tests	Result with the skip removed
1: vm.store on the PolicyRegistry precompile address	4: both counterOverflow tests, createPolicy_revertOrder, pendingPolicyAdmin storage poison	fail: forge rejects the store
2: mock-only slot	2: MockB20SlotHelpers	initializedSlot fails. adminCountSlot_disjoint passes, but it only compares library constants
3: unreachable bootstrap state	5: the privileged transferFrom and transferFromWithMemo tests	4 pass by accident: the store hits a slot that doesn't exist in Rust, so the window never reopens and the privileged path never runs. skipsExecutorPolicy fails with PolicyForbids, which confirms it
None: waiting on base/base#3362	2: B20FactoryNonPayable	both pass

The two B20FactoryNonPayable skips don't meet the criteria. base/base#3362 has merged, and both tests pass on Beryl and Denim. I left them alone to keep this PR to documentation; removing them is the follow-up.

Verification

forge fmt --check test/lib/BaseTest.sol and forge build pass, and the PolicyRegistryCreatePolicyTest and MockB20SlotHelpersTest suites pass in reference mode. The change is comments and docs only.

[github-actions]

Interface Coverage

✅ All interface functions have test coverage.

[github-actions]

📊 Forge Coverage (src/lib/)

🟡 ≥95% across all metrics — some metrics below 99%.

File	Lines	Stmts	Branches	Funcs
🟡 B20FactoryLib.sol	97.70%	98.00%	100.00%	95.00%
🔴 test/lib/ForceFeeder.sol	0.00%	0.00%	100.00%	0.00%
🔴 test/lib/PrecompileProbe.sol	0.00%	0.00%	0.00%	0.00%
🟢 MockActivationRegistry.sol	100.00%	100.00%	100.00%	100.00%
🟢 MockActivationRegistryStorage.sol	100.00%	100.00%	100.00%	100.00%
🟢 MockB20.sol	100.00%	99.68%	100.00%	100.00%
🟢 MockB20Asset.sol	100.00%	100.00%	100.00%	100.00%
🟡 MockB20Factory.sol	98.96%	99.10%	100.00%	100.00%
🟢 MockB20Stablecoin.sol	100.00%	100.00%	100.00%	100.00%
🟢 MockB20Storage.sol	100.00%	100.00%	100.00%	100.00%
🟢 MockBaseTime.sol	100.00%	100.00%	100.00%	100.00%
🟡 MockPolicyRegistry.sol	100.00%	99.58%	97.78%	100.00%
🟢 MockPolicyRegistryStorage.sol	100.00%	100.00%	100.00%	100.00%
Total	97.14%	97.51%	98.21%	97.07%

Full report: download artifact. To browse locally: make coverage (runs forge coverage + genhtml + opens the HTML report).

[github-actions]

❌ Fork tests did not complete

No completed, non-empty suite was recorded. Check the workflow logs for build, setup, or execution errors.