The Credible Layer consists of four key components working together:
Assertions: Security rules written in Solidity that define states that should never occur (e.g., “implementation address shouldn’t change”, “price reported by oracle shouldn’t deviate more than x% inside of a single transaction”).
Protocols: Protocols that define assertions for their contracts and link them on-chain.
Block Builders/Sequencers: The network infrastructure that validates every transaction against assertions before inclusion in a block, dropping any that would violate security rules.
Transparency Dashboard: A platform where users can see which protocols are protected and how, creating a focal point for ecosystem security.
OP-Talos receives the transaction for potential inclusion in a block
For each transaction, OP-Talos:
References the assertions stored in the Credible Layer Protocol that are related to the contracts the transaction interacts with
Fetches assertion bytecode from the Assertion DA
The PhEVM:
Simulate transaction execution
Create pre-transaction and post-transaction state snapshots
Execute all relevant assertions against these transactions
The PhEVM returns assertion results:
If any assertion reverts (invalid state), the transaction is flagged as invalid
If all assertions pass, the transaction is considered valid
OP-Talos makes inclusion decisions:
Invalid transactions are not included in the block
Valid transactions are included in the block
OP-Talos submits the validated block to the sequencer via Rollup Boost
Sequencer accepts and processes the validated block
Ultimately, if a transaction would result in a hack, if the contract is protected by the Credible Layer, the transaction will be dropped. If the contract is not protected, the transaction will be included in the block.
Now you’re probably thinking that forced inclusion transactions are not caught by the Credible Layer…
Well, think again! We have found a way to handle this and are actively working on implementing it.
Neutrality is critical. It determines who controls the system. Who controls the system decides how the system is regulated.
As middleware, the Credible Layer extends network capabilities without introducing new trust assumptions. Protocols define security rules through existing governance channels, and networks enforce these rules through existing mechanisms.
Networks can remove the Credible Layer anytime and still execute their core functionality. dApps can remove (or add) assertions anytime, and the system passes the off-switch test.
By avoiding centralized control points, this architecture preserves decentralization for both dApps and networks.
Policymakers are becoming increasingly aware of decentralization and see its value. Systems that quietly rely on centralized emergency controls for security risk falling on the wrong side of laws meant to capture decentralized-in-name-only projects.
Non-neutral solutions ultimately create new centralized control points: AI providers can change detection algorithms, and SaaS companies can modify security policies.
This matters because security solutions should optimally not add additional trust assumptions/risks, and they should not centralize the systems they serve.
Credible Layer dApp: User interface for protocols to register and manage assertions
Transparency Dashboard: The part of the dApp that shows every project’s assertions and other security measures to end users. These users can use these proofs to eliminate the blind trust that most use to make allocation decisions today.
pcl CLI: Command-line tool for writing, testing, and deploying assertions
credible-std: Standard library exposing cheatcodes and testing functionality for writing and testing assertions
Screenshot of the Credible Layer dApp UI for a project