brainwires-provider 0.11.0

LLM provider implementations (Anthropic, OpenAI Chat + Responses, Google Gemini, Ollama, Bedrock, Vertex AI, local llama.cpp) for the Brainwires Agent Framework. Speech (TTS/STT) providers live in `brainwires-provider-speech`.
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

//! OpenAI Responses API provider (`POST /v1/responses`).
//!
//! The Responses API is OpenAI's primary modern API surface, superseding Chat
//! Completions. Key differences:
//! - Input is `input` not `messages` — supports text string or structured items
//! - Output contains typed items (message, function_call, web_search, etc.)
//! - `previous_response_id` chains conversations server-side
//! - Supports built-in tools: web search, file search, code interpreter,
//!   computer use, MCP servers, image generation
//! - Streaming uses 30+ typed event types
//! - 6 REST endpoints: create, retrieve, delete, cancel, list_input_items, compact

pub mod client;
pub mod convert;
pub mod provider;
pub mod types;
pub mod websocket;
pub mod ws_provider;

// Re-export key public types at module level for ergonomic imports
pub use client::ResponsesClient;
pub use provider::OpenAiResponsesProvider;

// Re-export wire types used by external code
pub use types::{
    AudioOutputConfig, CreateResponseRequest, InputContent, InputContentPart, OutputContentBlock,
    ResponseInput, ResponseInputItem, ResponseObject, ResponseOutputItem, ResponseStreamEvent,
    ResponseTool, ResponseUsage, ToolChoice,
};

// WebSocket transport
pub use websocket::ResponsesWebSocket;
pub use ws_provider::OpenAiResponsesWsProvider;