Crate motosan_agent_loop 

motosan-agent-loop

Standalone ReAct agent loop — LlmClient + AgentLoop with no platform dependencies.

This crate provides the core loop that drives an LLM through iterative reasoning and tool execution, producing a final answer.

Re-exports

pub use context::ContextProvider;

pub use error::AgentError;

pub use error::Result;

pub use llm::ChatOutput;

pub use llm::LlmClient;

pub use llm::LlmResponse;

pub use llm::StreamChunk;

pub use llm::TokenUsage;

pub use llm::ToolCallItem;

pub use loop_::AgentEvent;

pub use loop_::AgentLoop;

pub use loop_::AgentLoopBuilder;

pub use loop_::AgentOp;

pub use loop_::AgentResult;

pub use loop_::BackpressurePolicy;

pub use loop_::ChannelConfig;

pub use message::Message;

pub use message::Role;

pub use message::ToolCallRef;

pub use session::AgentSession;

pub use session_store::FileSessionStore;

pub use session_store::MemorySessionStore;

pub use session_store::SessionMeta;

pub use session_store::SessionStore;

Modules

context

error

llm

loop_

message

session

Session lifecycle and persistence guarantees

session_store