praxis-graph 0.2.0

React agent orchestrator for building AI agent workflows with LLM and tool execution
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
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111

use crate::types::config::{LLMConfig, ContextPolicy};
use crate::types::GraphOutput;
use praxis_llm::{Message, ToolCall};
use serde::{Deserialize, Serialize};
use std::collections::HashMap;

#[derive(Debug, Clone)]
pub struct GraphState {
    pub conversation_id: String,
    pub run_id: String,
    pub messages: Vec<Message>,
    pub llm_config: LLMConfig,
    pub variables: HashMap<String, serde_json::Value>,
    #[allow(dead_code)]
    pub last_outputs: Option<Vec<GraphOutput>>,
}

impl GraphState {
    pub fn new(
        conversation_id: String,
        run_id: String,
        messages: Vec<Message>,
        llm_config: LLMConfig,
    ) -> Self {
        Self {
            conversation_id,
            run_id,
            messages,
            llm_config,
            variables: HashMap::new(),
            last_outputs: None,
        }
    }

    pub fn from_input(input: GraphInput) -> Self {
        Self {
            conversation_id: input.conversation_id,
            run_id: uuid::Uuid::new_v4().to_string(),
            messages: input.messages,
            llm_config: input.llm_config,
            variables: HashMap::new(),
            last_outputs: None,
        }
    }

    pub fn last_message(&self) -> Option<&Message> {
        self.messages.last()
    }

    pub fn add_message(&mut self, message: Message) {
        self.messages.push(message);
    }

    pub fn has_pending_tool_calls(&self) -> bool {
        if let Some(last_msg) = self.last_message() {
            match last_msg {
                Message::AI { tool_calls, .. } => tool_calls.is_some(),
                _ => false,
            }
        } else {
            false
        }
    }

    pub fn get_pending_tool_calls(&self) -> Vec<ToolCall> {
        if let Some(last_msg) = self.last_message() {
            match last_msg {
                Message::AI { tool_calls: Some(calls), .. } => calls.clone(),
                _ => Vec::new(),
            }
        } else {
            Vec::new()
        }
    }

    pub fn add_tool_result(&mut self, tool_call_id: String, result: String) {
        self.messages.push(Message::Tool {
            tool_call_id,
            content: praxis_llm::Content::text(result),
        });
    }
}

#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct GraphInput {
    pub conversation_id: String,
    pub messages: Vec<Message>,
    pub llm_config: LLMConfig,
    pub context_policy: ContextPolicy,
}

impl GraphInput {
    pub fn new(
        conversation_id: impl Into<String>,
        messages: Vec<Message>,
        llm_config: LLMConfig,
    ) -> Self {
        Self {
            conversation_id: conversation_id.into(),
            messages,
            llm_config,
            context_policy: ContextPolicy::default(),
        }
    }

    pub fn with_context_policy(mut self, policy: ContextPolicy) -> Self {
        self.context_policy = policy;
        self
    }
}