mermaid-cli 0.18.0

Open-source AI pair programmer with agentic capabilities. Local-first with Ollama, native tool calling, and beautiful TUI.
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

//! Provider-facing traits: `ModelProvider` + `ToolExecutor`.
//!
//! This module is the boundary between "inert data the reducer
//! produces" and "real I/O the effect runner performs." Concrete
//! provider adapters (Ollama, Anthropic, Gemini, OpenAI-compat) live
//! under `model/`; tool implementations live under `tool/`.
//!
//! The goal is that adding a provider or tool touches exactly one
//! file — no observer plumbing, no dispatch wiring, no adapter-
//! specific retry logic. The traits carry the full surface; the
//! effect runner just looks up an impl and calls it.

pub mod approval;
pub mod auto_classifier;
pub mod capabilities;
pub mod ctx;
pub mod factory;
pub mod model;
pub mod questions;
pub mod tasks;
pub mod tool;

pub use approval::{ApprovalBroker, ApprovalDecision, allowlist_key};
pub use auto_classifier::{AutoClassifier, ModelAutoClassifier, VetRequest, VetVerdict};
pub use capabilities::Capabilities;
pub use ctx::{
    ExecContext, FinalResponse, ProgressEvent, StreamContext, StreamEvent, SubagentPhase,
    clone_messages, test_exec_context, test_stream_context,
};
pub use factory::ProviderFactory;
pub use model::{
    AnthropicProvider, GeminiProvider, MetaProvider, ModelProvider, OllamaProvider,
    OpenAICompatProvider,
};
pub use questions::QuestionBroker;
pub use tasks::TaskBroker;
pub use tool::{ToolExecutor, ToolRegistry, TuiMode};

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn trait_is_object_safe() {
        // Compile-time guard: `dyn ModelProvider` must stay
        // object-safe so `ProviderFactory` can hand out
        // `Arc<dyn ModelProvider>`.
        fn _assert(_: &dyn ModelProvider) {}
    }
}