Skip to content

README.md

Latest commit

 

History

History
89 lines (64 loc) · 2.15 KB

README.md

File metadata and controls

89 lines (64 loc) · 2.15 KB
Raw

Value SDK Examples

Code examples demonstrating how to instrument AI agents with the Value SDK.

Python Examples

Example Description
basic_sync.py Synchronous SDK usage
basic_async.py Async SDK usage
gemini_instrumentation.py Auto-instrument Gemini API calls
pipeline_simulation.py Complete AI pipeline simulation
otel_basics.py Raw OpenTelemetry tracing

Prerequisites

  1. Install the Value SDK:

    pip install value-sdk

  2. Set environment variables:

    export VALUE_SECRET=your-agent-secret
export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318

  3. Start the platform (if running locally):

    make value-up

Running Examples

cd examples/python

# Basic examples
python basic_sync.py
python basic_async.py

# Gemini auto-instrumentation (requires GOOGLE_API_KEY)
export GOOGLE_API_KEY=your-key
python gemini_instrumentation.py

# Full pipeline simulation (runs continuously)
python pipeline_simulation.py

SDK Quick Reference

Initialize Client

# Sync
from value import initialize_sync
client = initialize_sync()

# Async
from value import initialize_async
client = await initialize_async()

Send Actions

with client.action_context(user_id="user123", anonymous_id="anon456") as ctx:
    ctx.send(
        action_name="my_action",
        **{
            "value.action.description": "Action description",
            "custom_attribute": "value",
        },
    )

Auto-Instrument LLMs

from value.instrumentation import auto_instrument

auto_instrument(["gemini"])  # Supported: gemini, openai, anthropic

License

MIT License — See LICENSE