layer0 0.4.0

Protocol traits for composable agentic AI systems
Documentation

layer0

Protocol traits for composable agentic AI systems

crates.io docs.rs license

Overview

layer0 defines the foundational protocol traits that the entire neuron workspace builds on. It contains no implementations — only the contracts that every agentic component must satisfy.

The traits here form the 6-layer model:

Layer Trait(s) Responsibility
0 Operator, Effect Invoke operators, emit structured effects
1 StateStore Persistent key-value memory
2 Environment, CredentialRef Credential injection + env config
3 Orchestrator, WorkflowHandle Coordinate multi-operator workflows
4 Hook, HookRegistry Pre/post lifecycle middleware
5 Observable Structured observability events

Usage

[dependencies]
layer0 = "0.4"

Test utilities

[dev-dependencies]
layer0 = { version = "0.4", features = ["test-utils"] }

The test-utils feature exports stub implementations (StubOperator, StubStateStore, etc.) useful for testing downstream crates without pulling in full implementations.

Part of the neuron workspace

neuron is a composable async agentic AI framework for Rust. See the book for architecture and guides.