Golf v0.2.0 introduces powerful built-in utilities that enhance your tool development experience. These utilities are automatically available in all tools without additional imports and provide seamless integration with the Golf Platform for enhanced telemetry and monitoring.

​
Overview

Golf utilities provide three core capabilities:
  1. Elicitation - Interactive user prompting and input collection
  2. Sampling - LLM interactions and text generation
  3. Context Management - Access to context and request information
All utilities respect your authentication configuration and provide automatic telemetry when used with the Golf Platform.

​
Elicitation Utilities

Elicitation utilities enable your tools to interact with users, collect additional information, and create dynamic conversational experiences.

​
Basic Usage

from golf.utils import elicit

async def my_interactive_tool():
    # Simple text prompt
    color = await elicit("What's your favorite color?")
    
    # Use the response in your tool logic
    return {"message": f"Great choice! {color} is a beautiful color."}

​
Advanced Elicitation

from golf.utils import elicit

async def configuration_tool():
    # Collect multiple pieces of information
    name = await elicit("What's your name?")
    email = await elicit("What's your email address?")
    preferences = await elicit("Any special preferences? (optional)")
    
    # Process and use the collected data
    config = {
        "name": name,
        "email": email,
        "preferences": preferences or "None specified"
    }
    
    return {"configuration": config}

​
Elicitation with Validation

from golf.utils import elicit

async def age_verification_tool():
    while True:
        age_str = await elicit("Please enter your age:")
        try:
            age = int(age_str)
            if age >= 18:
                break
            else:
                await elicit("You must be 18 or older. Please try again.")
        except ValueError:
            await elicit("Please enter a valid number.")
    
    return {"verified": True, "age": age}

​
Sampling Utilities

Sampling utilities provide direct access to LLM capabilities within your tools, enabling text generation, analysis, and AI-powered processing.

​
Basic Sampling

from golf.utils import sample

async def explanation_tool(topic: str):
    # Generate an explanation using the default model
    explanation = await sample(f"Explain {topic} in simple terms")
    
    return {"topic": topic, "explanation": explanation}

​
Model-Specific Sampling

from golf.utils import sample

async def analysis_tool(text: str):
    # Use a specific model for analysis
    sentiment = await sample(
        f"Analyze the sentiment of this text: '{text}'. Respond with only: positive, negative, or neutral.",
        model="gpt-4"
    )
    
    summary = await sample(
        f"Summarize this text in one sentence: '{text}'",
        model="gpt-3.5-turbo"
    )
    
    return {
        "original_text": text,
        "sentiment": sentiment.strip(),
        "summary": summary
    }

​
Advanced Sampling with Parameters

from golf.utils import sample

async def creative_writing_tool(prompt: str, style: str):
    # Advanced sampling with custom parameters
    story = await sample(
        f"Write a short story in {style} style based on: {prompt}",
        model="gpt-4",
        max_tokens=500,
        temperature=0.8
    )
    
    return {
        "prompt": prompt,
        "style": style,
        "story": story
    }

​
Context Management

Context utilities provide access to context, request information, and authentication details.

​
Basic Context Access

from golf.utils import get_context

async def context_aware_tool():
    # Get the current context
    context = get_context()
    
    # Access request information
    request_id = context.get("request_id")
    client_info = context.get("client_info")
    
    return {
        "request_id": request_id,
        "client": client_info
    }

​
Authentication Context

from golf.utils import get_context
from golf.auth import get_api_key

async def authenticated_tool():
    # Get context and auth information
    context = get_context()
    api_key = get_api_key()
    
    # Use context for logging or customization
    user_id = context.get("user_id")
    session_id = context.get("session_id")
    
    return {
        "authenticated": bool(api_key),
        "user_id": user_id,
        "session_id": session_id
    }

​
Request Metadata

from golf.utils import get_context

async def metadata_tool():
    context = get_context()
    
    # Access various metadata
    metadata = {
        "timestamp": context.get("timestamp"),
        "client_version": context.get("client_version"),
        "transport": context.get("transport"),
        "headers": context.get("headers", {})
    }
    
    return {"request_metadata": metadata}