ares-server 0.6.2

A.R.E.S - Agentic Retrieval Enhanced Server: A production-grade agentic chatbot server with multi-provider LLM support, tool calling, RAG, and MCP integration
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

use crate::{
    auth::middleware::AuthUser,
    research::coordinator::ResearchCoordinator,
    types::{ResearchRequest, ResearchResponse, Result},
    AppState,
};
use axum::{extract::State, Json};
use std::time::Instant;

/// Perform deep research on a query
#[utoipa::path(
    post,
    path = "/api/research",
    request_body = ResearchRequest,
    responses(
        (status = 200, description = "Research completed", body = ResearchResponse),
        (status = 400, description = "Invalid input"),
        (status = 401, description = "Unauthorized")
    ),
    tag = "research",
    security(("bearer" = []))
)]
pub async fn deep_research(
    State(state): State<AppState>,
    AuthUser(_claims): AuthUser,
    Json(payload): Json<ResearchRequest>,
) -> Result<Json<ResearchResponse>> {
    let start = Instant::now();

    // Get research workflow config
    let config = state.config_manager.config();
    let (depth, max_iterations) = if let Some(workflow) = config.get_workflow("research") {
        (
            payload.depth.unwrap_or(workflow.max_depth),
            payload.max_iterations.unwrap_or(workflow.max_iterations),
        )
    } else {
        (
            payload.depth.unwrap_or(2),
            payload.max_iterations.unwrap_or(5),
        )
    };

    // Get model for orchestrator (used for research)
    let model_name = config
        .get_agent("orchestrator")
        .map(|a| a.model.as_str())
        .unwrap_or("powerful");

    // Create research coordinator with config-based model
    let llm_client = match state
        .provider_registry
        .create_client_for_model(model_name)
        .await
    {
        Ok(client) => client,
        Err(_) => state.llm_factory.create_default().await?,
    };

    let coordinator = ResearchCoordinator::new(llm_client, depth, max_iterations);

    // Execute research
    let (findings, sources) = coordinator.research(&payload.query).await?;

    let duration = start.elapsed();

    Ok(Json(ResearchResponse {
        findings,
        sources,
        duration_ms: duration.as_millis() as u64,
    }))
}