formal-ai 0.302.0

Formal symbolic AI implementation with OpenAI-compatible APIs
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

//! Complete agentic conversation export and learning endpoints (#822).

use serde_json::json;

use super::{error_response, json_response, links_notation_response, query_param, ApiHttpResponse};
use crate::memory_sync::SyncStore;

const ERROR_PLACEHOLDER: &str = "{error}";

pub(super) fn handle_context_request(dialog_id: &str, query: &str) -> ApiHttpResponse {
    let mut context = match crate::conversation_context::load_conversation_context(dialog_id) {
        Ok(context) => context,
        Err(error) if error.kind() == std::io::ErrorKind::InvalidInput => {
            return error_response(400, &error.to_string());
        }
        Err(error) if error.kind() == std::io::ErrorKind::NotFound => {
            return error_response(404, &error.to_string());
        }
        Err(error) => return error_response(500, &error.to_string()),
    };
    match query_param(query, "include").as_deref() {
        Some("harness") => {
            if let Some(object) = context.as_object_mut() {
                object.remove("server_logs");
            }
        }
        Some("server") => {
            if let Some(object) = context.as_object_mut() {
                object.remove("messages");
            }
        }
        Some("both") | None => {}
        Some(_) => return error_response(400, "include must be harness, server, or both"),
    }
    if query_param(query, "format").as_deref() == Some("json") {
        return json_response(200, &context);
    }
    links_notation_response(
        200,
        crate::conversation_context::conversation_context_to_lino(dialog_id, &context),
    )
}

pub(super) fn handle_learning_request(dialog_id: &str) -> ApiHttpResponse {
    let context = match crate::conversation_context::load_conversation_context(dialog_id) {
        Ok(context) => context,
        Err(error) if error.kind() == std::io::ErrorKind::InvalidInput => {
            return error_response(400, &error.to_string());
        }
        Err(error) if error.kind() == std::io::ErrorKind::NotFound => {
            return error_response(404, &error.to_string());
        }
        Err(error) => return error_response(500, &error.to_string()),
    };
    let document = crate::conversation_context::conversation_context_to_lino(dialog_id, &context);
    let staged = crate::self_improvement::learn_from_reported_conversation(&context);
    let mut store = SyncStore::open();
    match store.record_chat_exchange(&format!("agentic_report_{dialog_id}"), &document) {
        Ok(events_recorded) => json_response(
            200,
            &json!({
                "dialog_id": dialog_id,
                "learned": true,
                "events_recorded": events_recorded,
                "learning_trace_found": staged.is_some(),
                "rule_proposals": staged.as_ref().map_or(0, |run| run.learning.proposals.len()),
                "awaiting_human_review": staged.as_ref().is_some_and(|run| run.awaiting_human_review),
                "promoted": false,
            }),
        ),
        Err(error) => error_response(
            500,
            &config("context_learning_failed").replace(ERROR_PLACEHOLDER, &error.to_string()),
        ),
    }
}

fn config(key: &str) -> String {
    crate::seed::agent_info()
        .remove(key)
        .unwrap_or_else(|| key.to_owned())
}