​
Getting Started with Praxos

This quickstart guide will help you integrate and utilize the Praxos effectively. You’ll learn how to:
  • Configure your development environment
  • Initialize the Praxos client
  • Create and manage environments
  • Add conversations to the knowledge graph
  • Retrieve contextual information

​
Prerequisites

Before you begin, ensure you have:
  • A valid Praxos API key (see API Key Instructions)
  • Python 3.6 or higher installed
  • Basic understanding of Python programming

​
Environment Setup

  1. Install the Praxos Python SDK:
pip install -U praxos-python
  1. Configure your API key:
import requests

# Replace with your Praxos API key
API_KEY = "YOUR_API_KEY"

​
Initialize the Client

Create a Praxos client instance and initialize an environment: 
from praxos_python import SyncClient

# Initialize the client with your API key
client = SyncClient(api_key=API_KEY)

# Create a new environment
env = client.create_environment(name="MyEnvironment")
To learn more about client please review this page.

​
Add Conversations to Knowledge Graph

Add LLM conversations to the Praxos knowledge graph: 
from raxos_python.types.message import Message

messages = [
    Message(role="user", content="Hello, I am using Praxos knowledge graph, and it's awesome."),
    Message(role="assistant", content="Hi there! I'm glad you're exploring Praxos knowledge graph. It's a powerful tool for managing and retrieving contextual information.")
]

source = env.add_conversation(
    messages=messages,
    source_name="My Conversation",  # Optional identifier
    description="A sample conversation"  # Optional description
)
To learn more about environment please review this page.

​
Monitor Indexing Status

The knowledge graph requires time to index new sources. Indexing duration varies based on source size. Ensure indexing is complete before querying: 
import time

# Wait for indexing to complete
while source.get_status() != "success":
    time.sleep(1)
More document about source is available here

​
Retrieve Contextual Information

Query the knowledge graph to retrieve relevant context: 
# Query the knowledge graph
context = env.get_context("what is the name of knowledge graph tool?")
print(context.sentence)
Example output: 
a KnowledgeGraphTool, (which is a Tool), (category: Physical), identified as '67B770787F48A6Eac8C8B578 683A42Bb3983E0Bef0E2D031 683A42Bb3983E0Bef0E2D030 0 Tool Praxoskg', which has attribute ToolName 'Praxos knowledge graph' (base: NameType), and has attribute ToolFunctionalityDescription 'powerful tool for managing and retrieving contextual information' (base: DescriptionType). has attribute ToolName 'Praxos knowledge graph' (base: NameType).
The retrieved context can be utilized to enhance LLM responses for your specific use case.