Module processing 

Request-processing pipeline for the A2A server.

Turns an inbound A2A Message into a persisted Task: validation, conversation-history assembly, LLM execution (via ExecutionStrategy), artifact construction, response synthesis, and persistence. The entry point is MessageProcessor; StreamProcessor drives the streaming variant and emits StreamEvents.

Copyright (c) systemprompt.io — Business Source License 1.1. See https://systemprompt.io for licensing details.

Re-exports

pub use artifact::ArtifactBuilder;

pub use conversation_service::ConversationService;

pub use message::MessageProcessor;

pub use message::StreamEvent;

pub use message::StreamProcessor;

pub use persistence_service::PersistenceService;

pub use strategies::ExecutionContext;

pub use strategies::ExecutionResult;

pub use strategies::ExecutionStrategy;

pub use strategies::ExecutionStrategySelector;

pub use task_builder::TaskBuilder;

pub use task_builder::build_canceled_task;

pub use task_builder::build_completed_task;

pub use task_builder::build_mock_task;

pub use task_builder::build_multiturn_task;

Modules

ai_executor

Direct AI-provider calls for the processing pipeline.

artifact

Construction of A2A Artifacts from MCP tool results.

conversation_service

Conversation-history assembly for AI requests.

message

Message processing for the A2A server.

persistence_service

Task persistence for the processing pipeline.

strategies

Execution strategies for turning AI messages into a response.

task_builder

Assembly of A2A Task values for the various completion paths.