mindfork 0.11.0

A terminal AI chat written in Rust: local models via llama.cpp or OpenAI, Anthropic, Gemini and Grok in the cloud, with persistent memory, notes, RAG and tools.
1
2
3
4
5
6
7
8
9
10
11
12
13

//! OpenAI cloud backend over the **Responses API** (`POST /v1/responses`). A protocol
//! separate from Chat Completions, same vendor (ADR 0004,
//! docs/research/openai-responses-client.md): reasoning summaries
//! (`reasoning.summary`), reasoning depth (`reasoning.effort`), verbosity
//! (`text.verbosity`), reasoning items with `encrypted_content` for tool-use.
//!
//! `wire` — building the request body and parsing event-based SSE; `client` — the
//! [`EngineBackend`](crate::shared::api::contract::EngineBackend) implementation.

mod client;
mod wire;

pub use client::ResponsesClient;