Integration: LangWatch

Trace your Haystack pipelines and agents with LangWatch

Authors

LangWatch

Table of Contents

• Overview
• Installation
• Usage
• License

Overview

LangWatch is an open source LLM observability platform. LangWatch integrates with Haystack through OpenInference instrumentation to capture traces from your Haystack pipelines and components, including:

• Pipeline operations
• Component executions
• Model interactions

You can learn more about the Haystack and LangWatch integration in LangWatch’s Haystack integration guide.

Installation

To use the LangWatch integration with Haystack, install the langwatch package together with the OpenInference Haystack instrumentation:

pip install langwatch openinference-instrumentation-haystack haystack-ai

Usage

To use LangWatch, you will need to:

1. Set your LangWatch API key in the LANGWATCH_API_KEY environment variable
2. Pass HaystackInstrumentor to langwatch.setup()

An example agent traced with LangWatch is shown below:

import os
import langwatch

from haystack.components.agents import Agent
from haystack.components.generators.chat import OpenAIChatGenerator
from haystack.dataclasses import ChatMessage
from openinference.instrumentation.haystack import HaystackInstrumentor

langwatch.setup(instrumentors=[HaystackInstrumentor()])

basic_agent = Agent(
    chat_generator=OpenAIChatGenerator(model="gpt-5-mini"),
    system_prompt="You are a helpful web agent.",
    tools=[],
)

result = basic_agent.run(messages=[ChatMessage.from_user("Tell me a joke")])

print(result["last_message"].text)

The HaystackInstrumentor automatically captures Haystack pipeline operations, component executions, and model interactions. Use @langwatch.trace() to create a parent trace under which Haystack operations will be nested.

License

LangWatch is open source and is distributed under the terms of the Apache-2.0 license.