Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.xandrlabs.ai/llms.txt

Use this file to discover all available pages before exploring further.

ALX Protocol is an open protocol that gives you a standard way to create, identify, link, and verify structured artifacts — called Knowledge Blocks — across any system, language, or runtime. It defines how blocks are canonicalized, how their identities are derived, how their relationships form a verifiable lineage graph, and how value flows to contributors through on-chain attribution. The protocol does not provide a hosted service or a product UI; it is infrastructure that you integrate.

​
Core concepts

Knowledge Blocks are the fundamental unit of ALX Protocol. A Knowledge Block (KB) is an immutable, typed artifact defined by its content, identity, dependencies, and provenance. KBs can represent solutions, standards, patterns, evaluation rubrics, compliance checklists, and any other form of structured knowledge you want to track and reuse. Every Knowledge Block has four properties:
  • Identity — a deterministic kbHash derived by canonicalizing the block’s envelope (using RFC 8785 / JCS) and hashing it with a domain-separated keccak256. Identical blocks always produce identical identifiers.
  • Lineage — explicit references to parent blocks form a directed acyclic graph (DAG). Derivation relationships are tracked, validated for cycles, and sorted deterministically before hashing.
  • Conformance — independent implementations must produce identical outputs for identical inputs. Correctness is defined by canonical test vectors, not by any single runtime.
  • Attribution and settlement — query fees flow through the lineage DAG to contributors proportionally, enforced by smart contracts on Base mainnet.

​
Key features

Deterministic identity

The same content always produces the same kbHash. There is no server-assigned ID — identity is a property of the content itself.

Lineage DAG

Knowledge Blocks reference parent blocks explicitly. The resulting DAG is the canonical record of how artifacts are composed and derived.

On-chain attribution

The registry contract on Base mainnet records every published block and enforces royalty distribution across the lineage graph during settlement.

IPFS pinning

Artifacts are addressed by their IPFS CID, giving you content-addressed storage that is independent of any single host or service.

​
What you can build

ALX Protocol is suited for any system that requires correctness, reproducibility, and traceability.
  • AI pipelines — produce consistent, reproducible outputs across runs and implementations by anchoring every artifact to its deterministic identity.
  • Audit systems — verify that a retrieved artifact matches its published cryptographic commitment and trace it back through its full input lineage.
  • Multi-agent coordination — establish shared, deterministic state that independent agents can agree on without a central coordinator.
  • Artifact systems — compose, reuse, and standardize verified knowledge without duplication or drift across teams.

​
On-chain deployment

The ALX Protocol registry is deployed on Base mainnet at address 0xD1F216E872a9ed4b90E364825869c2F377155B29 (chain ID 8453). Publishing a Knowledge Block requires a minimum stake of 0.001 ETH. Artifacts are pinned to IPFS and referenced by CID.

​
Supported languages

The TypeScript package @alx/protocol is the normative reference implementation. Community implementations are also available for Python and Rust. All conformant implementations must produce identical outputs for identical inputs — the test vectors are the source of truth.

​
Where to go next

Quick start

Compute your first Knowledge Block identity in minutes.

Knowledge Blocks

Understand the structure of a KB envelope and how identity is derived.

Compute identity

Step-by-step guide to canonicalization and kbHash derivation.

API reference

Full reference for every function exported by @alx/protocol.