systemprompt-agent 0.47.0

Agent-to-Agent (A2A) protocol for systemprompt.io AI governance: streaming, JSON-RPC models, task lifecycle, .well-known discovery, and governed agent orchestration.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73

//! # systemprompt-agent
//!
//! Agent-to-Agent (A2A) protocol implementation for systemprompt.io.
//!
//! This crate hosts the domain logic for hosting governed AI agents:
//! - JSON-RPC protocol models in [`models::a2a`]
//! - Persistence in [`repository`] (tasks, contexts, artifacts, execution
//!   steps)
//! - Runtime services in [`services`] (HTTP server, orchestration, MCP
//!   bridging, skills ingestion, streaming, registry)
//! - A typed error hierarchy rooted at [`AgentError`]
//!
//! ## Feature flags
//!
//! This crate currently has no Cargo features beyond defaults. Functionality
//! is unconditional and the crate compiles a single configuration. The facade
//! crate `systemprompt` gates inclusion via the `agent`/`full` features.
//!
//! ## Layer
//!
//! Domain layer. Depends only on `shared/*` and `infra/*` crates plus a small
//! number of sibling domain crates (declared in `Cargo.toml`).
//!
//! Copyright (c) systemprompt.io — Business Source License 1.1.
//! See <https://systemprompt.io> for licensing details.

/// Internal seams exposed for the out-of-tree test workspace.
///
/// Not part of the semver-stable surface. These items are crate-private
/// because callers must enter through the message-processing pipeline, which
/// sequences them against the task lifecycle; the wrappers here delegate
/// without altering that behaviour.
#[doc(hidden)]
pub mod test_api {
    pub use crate::services::a2a_server::processing::message::message_handler::{
        new_submitted_task, resolve_agent_message, resolve_task_id,
    };
    pub use crate::services::a2a_server::processing::message::persistence::{
        PersistCompletedTaskParams, broadcast_completion, persist_completed_task,
    };
    pub use crate::services::a2a_server::processing::message::stream_processor::helpers::build_artifacts_from_results;
}

pub(crate) mod error;
pub(crate) mod extension;
pub mod models;
pub mod repository;
pub mod services;
pub(crate) mod state;

pub use extension::AgentExtension;

pub use state::AgentState;

pub use models::a2a::{
    A2aJsonRpcRequest, A2aRequestParams, A2aResponse, AgentCapabilities, AgentCard, AgentInterface,
    AgentProvider, AgentSkill, Artifact, DataPart, Message, MessageSendParams, Part,
    SecurityScheme, Task, TaskIdParams, TaskQueryParams, TaskState, TaskStatus, TextPart,
    TransportProtocol,
};

pub use error::{
    AgentError, AgentResult, ArtifactError, ContextError, ProtocolError, RowParseError, TaskError,
};

pub const A2A_PROTOCOL_VERSION: &str = "0.3.0";

pub use services::{
    AgentEvent, AgentEventBus, AgentHandlerState, AgentOrchestrator, AgentServer, AgentStatus,
    ContextService, SkillService,
};

pub use repository::content::ArtifactRepository;