Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.cognee.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Observability with Langfuse

Langfuse is an open-source observability and analytics platform for LLM-powered applications. It collects traces, generations, and custom metrics so that you can debug, evaluate, and monitor your AI features in production.

​
Integration inside Cognee

Cognee ships with Langfuse support as an optional dependency. The integration is intentionally lightweight and consists of just a few lines of code. Anywhere in the codebase where we want observability we add: 
from cognee.modules.observability.get_observe import get_observe
observe = get_observe()

@observe(as_type="generation")  # optional label
async def acreate_structured_output(...):
    ...
The decorator is a thin wrapper around Langfuse and automatically creates a span every time the function is executed.
The data is sent to the Langfuse backend specified via environment variables and can be inspected in the Langfuse UI.

​
Quick Start

​
1. Install cognee with monitoring support

Langfuse is included as part of cognee’s monitoring optional dependencies. You need to install cognee with the monitoring extra: 
pip install 'cognee[monitoring]'
Cognee currently supports langfuse>=2.32.0,<3.
Running from the cognee source repository? If you’re developing with the cognee core repo, run the following from the project root on the main branch: 
uv sync --all-extras
This will install all optional dependencies, including monitoring/Langfuse support.

​
2. Create a Langfuse project

Create a project at Langfuse Cloud and copy the public and secret keys.

​
3. Configure environment variables

Export the following environment variables (for example in your .env file): 
LANGFUSE_PUBLIC_KEY=<your public key>
LANGFUSE_SECRET_KEY=<your secret key>
LANGFUSE_HOST=https://cloud.langfuse.com
BaseConfig reads these values on startup, so nothing else is required. Start Cognee as usual and open the Langfuse UI – traces will appear in real time.

​
4. Run your regular Cognee workflows

import cognee
import asyncio
from cognee.modules.observability.get_observe import get_observe

observe = get_observe()

@observe(name="simple_example_run", as_type="example")
async def main():
    await cognee.add("Natural language processing (NLP) is ...")
    await cognee.cognify()
    results = await cognee.search("Tell me about NLP")
    for r in results:
        print(r)

asyncio.run(main())

​
Adding your own spans

You can instrument any function in your codebase: 
from cognee.modules.observability.get_observe import get_observe
observe = get_observe()

@observe(as_type="my_tool", metadata={"foo": "bar"})
def my_helper(arg1, arg2):
    ...
You can pass the same keyword arguments that Langfuse’s observe() decorator accepts. With Langfuse v2 (cognee[monitoring] pins langfuse>=2.32.0,<3), common options include:
KwargDescription
nameOverride the span name (defaults to the function name)
as_typeSpan type label, e.g. "generation"
capture_inputWhether to capture function arguments
capture_outputWhether to capture the return value
metadataArbitrary dict attached to the span
Some integrations use @observe(workflow=True), but that is not part of Cognee’s built-in Langfuse integration. Cognee passes decorator kwargs through to Langfuse as-is, so unsupported arguments will fail at runtime.
TypeError: LangfuseDecorator.observe() got an unexpected keyword argument 'workflow'
If you see this error, remove workflow=True and use only the arguments supported by your installed Langfuse SDK version.
See the Langfuse Python SDK docs for the full reference.

​
OpenTelemetry tracing

Cognee also ships a standalone OpenTelemetry tracing layer that works independently of Langfuse and can export spans to Grafana, Jaeger, Dash0, and any other OTLP-compatible backend. See the OpenTelemetry Tracing page for setup details.
Join Langfuse and cognee communities on Discord for any questions.