Concepts
These pages explain the reasoning behind the system’s design — not how to accomplish a task (see How-to guides) and not exact technical specifications (see Reference).
Foundations
Section titled “Foundations” CKB cell model and lock scripts The CKB concepts you need before the rest of the docs make sense — cells, locks, cell deps, Type ID.
Why this exists The threat this solves: compromised runtimes, prompt injection, and why application-layer checks are not enough.
The two-layer model Why both the SDK check and the on-chain lock are necessary, and what each one actually guarantees.
System design
Section titled “System design” Governance design Why proposals need on-chain anchors, why the review window is at consensus, and why the validator Merkle tree.
Registry and cell design Why a single consumed-and-recreated cell, why sorted entries, and what fail-closed means operationally.
The autonomous treasury model Why an autonomous treasury instead of a funded signer, and how keyless spending works.
For adopters and auditors
Section titled “For adopters and auditors” Security model What the system protects, what it explicitly does not, and the registry serialization constraint.
Before you adopt the firewall lock Irreversible operational consequences — address migration, immutable lock args, governance timing.
Trust model and guarantees Exact guarantees, trust assumptions, what governance key compromise means, and what is auditable on-chain.