bob-core 0.2.2

Core domain types and port traits for Bob Agent Framework
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
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115

//! # Bob Core
//!
//! Domain types and port traits for the [Bob Agent Framework](https://github.com/longcipher/bob).
//!
//! ## Overview
//!
//! This crate defines the hexagonal boundary of the Bob Agent Framework using the
//! ports and adapters (hexagonal) architecture pattern. It contains:
//!
//! - **Domain Types**: Core data structures used throughout the framework
//! - **Port Traits**: Abstract interfaces that adapters must implement
//! - **Error Types**: Comprehensive error definitions for all components
//!
//! **No concrete implementations live here** — only contracts.
//!
//! ## Architecture
//!
//! The crate defines four primary port traits:
//!
//! 1. [`ports::LlmPort`] - Interface for language model interactions
//! 2. [`ports::ToolPort`] - Interface for tool/system operations
//! 3. [`ports::SessionStore`] - Interface for session state persistence
//! 4. [`ports::EventSink`] - Interface for event observation and logging
//!
//! ## Example
//!
//! ```rust,ignore
//! use bob_core::{
//!     ports::LlmPort,
//!     types::{LlmRequest, LlmResponse},
//!     error::LlmError,
//! };
//! use async_trait::async_trait;
//!
//! struct MyCustomLlm;
//!
//! #[async_trait]
//! impl LlmPort for MyCustomLlm {
//!     async fn complete(&self, req: LlmRequest) -> Result<LlmResponse, LlmError> {
//!         // Your implementation here
//!         todo!("implement LLM completion")
//!     }
//!
//!     async fn complete_stream(&self, req: LlmRequest) -> Result<LlmStream, LlmError> {
//!         // Your implementation here
//!         todo!("implement streaming completion")
//!     }
//! }
//! ```
//!
//! ## Features
//!
//! - **Zero-cost abstractions**: All traits use `async_trait` for async support
//! - **Type-safe**: Strong typing throughout with comprehensive error handling
//! - **Serializable**: All domain types implement `serde::Serialize` and `Deserialize`
//! - **Thread-safe**: All traits require `Send + Sync`
//!
//! ## Related Crates
//!
//! - [`bob_runtime`] - Runtime orchestration layer
//! - [`bob_adapters`] - Concrete adapter implementations
//!
//! [`bob_runtime`]: https://docs.rs/bob-runtime
//! [`bob_adapters`]: https://docs.rs/bob-adapters

#![deny(unsafe_code)]

// ── Modules ──────────────────────────────────────────────────────────

pub mod channel;
pub mod character;
pub mod error;
pub mod error_classifier;
pub mod instrumenter;
pub mod journal;
pub mod native;
pub mod ports;
pub mod resilience;
pub mod secrets;
pub mod tape;
pub mod tool_policy;
pub mod typed_tool;
pub mod types;

// ── Re-exports ───────────────────────────────────────────────────────

pub use character::Character;
pub use error::{AgentError, CostError, LlmError, StoreError, ToolError};
pub use error_classifier::{
    FailoverConfig, FailoverResult, classify_agent_error, classify_llm_error,
};
pub use instrumenter::{
    Instrumenter, ModelErrorInfo, ModelRequestInfo, ModelResponseInfo, NoopInstrumenter,
    OutputValidationErrorInfo, RunEndInfo, RunErrorInfo, RunStartInfo, ToolCallInfo,
    ToolDiscoveredInfo, ToolEndInfo, ToolErrorInfo,
};
pub use journal::{JournalEntry, ToolJournalPort};
pub use native::{NativeLlmPort, NativeSessionStore, NativeToolPort};
pub use ports::{
    AccessControlPort, ActivityJournalPort, ApprovalPort, ArtifactStorePort, CostMeterPort,
    EventSink, LlmPort, MessageBusPort, SessionStore, SubagentPort, TapeStorePort, ToolCatalogPort,
    ToolExecutorPort, ToolPolicyPort, ToolPort, TurnCheckpointStorePort,
};
pub use resilience::{
    CircuitBreaker, CircuitBreakerConfig, CircuitBreakerError, CircuitState, ComponentHealth,
    HealthStatus, RetryConfig,
};
pub use tool_policy::{
    intersect_allowlists, is_tool_allowed, merge_allowlists, normalize_tool_list, tools_match,
};
pub use typed_tool::{
    ChainedToolPorts, Complete, Described, FilteredToolPort, Incomplete, ToolKind, ToolPortExt,
    ToolSource, TypedToolAdapter, TypedToolBuilder, TypedToolExt, TypedToolPort,
};
pub use types::*;