Skip to main content
On Ethereum, the Assertion Enforcer reports failures to the participating builder, which excludes failing candidates before inclusion. The enforcer does not make the inclusion decision. A separate on-chain guard can revert a protected operation in a noncredible block while its gate is closed. In local CredibleTest tests, failed assertions cause reverts. This lets you use vm.expectRevert() to verify your assertion catches violations and returns the expected error message.

Quick Comparison

Production: Ethereum Builders Apply the Result

In production, the Assertion Enforcer enforces assertions on transactions during block building: The candidate must pass its matching active Production assertions. The builder excludes failing candidates. Satisfying the enforced assertions does not guarantee inclusion. Guards restrict protected operations in other builders’ blocks.

Testing: Transactions Revert

In local tests, cl.assertion() registers an assertion to run on the next external call: Key behaviors:
  • You specify the assertion function via fnSelector, but triggers still determine if it runs
  • Only the specified assertion runs, not all assertions for the contract
  • Assertions are enforced on only the next external call, then the registration is consumed
  • Reverts simulate drops - use vm.expectRevert() to test failure cases

The cl.assertion() Mental Model

cl.assertion() selects one assertion function for the next monitored external call. An intervening external call consumes the registration. The selected function still needs a matching trigger; selecting it is not a substitute for registering the trigger. The test wrapper models assertion rejection as a revert so a test can check both the reason and rollback. It does not reproduce Ethereum builder selection, credible-block marking, or an onchain guard decision. See How to Test Assertions for the setup pattern.

State Persistence

A passing operation retains its state changes for later calls in the test. A rejected operation rolls back. Each subsequent monitored call needs its own registration. To test several operations as one transaction, use a batch helper.

Common Pitfalls

”Expected 1 assertion to be executed, but 0 were executed”

This local-harness message means the expected function did not execute. It does not demonstrate that the policy accepted or rejected the operation. The troubleshooting guide covers consumed registrations, trigger mismatches, and protocol reverts.

Trigger Mismatch

A test can select the right assertion function but call an operation outside that function’s registered triggers. A zero-execution result identifies a coverage or test-setup gap. Changing the assertion condition cannot fix a trigger that never selects it; compare the target selector and registration in Triggers.

Alerting and Backtesting

Alerting evaluates mirrored live activity without excluding live transactions. Release backtesting evaluates a reported historical range under the proposed release. Neither uses the local test’s next-call registration, and neither proves that Production guards or routing work. A clean result needs evidence that relevant transactions exercised the assertions. Use Backtesting to review historical coverage and the integration test plan for the remaining enforcement checks.

Testing Assertions

Testing patterns

Triggers

Production trigger behavior

Backtesting

Review historical findings attached to a platform release

Troubleshooting

Common errors