​
Unlock LLM Observability

Agensight is an Observability SDK built for developers working with Large Language Models. It empowers you to trace every interaction, debug effectively, and optimize your prompts and LLM-powered applications with ease, all within your local development environment.

Quickstart Guide

Get Agensight installed and running your first trace in minutes.

Explore Core Features

Discover the power of automated tracing, local debugging, and the prompt playground.

​
Key Capabilities

Agensight provides a suite of tools designed to give you deep insights into your LLM workflows:

Seamless Tracing

Automatically capture structured traces for every LLM call, complete with context and token usage.

Local-First Debugging

Inspect traces, manage prompt versions, and experiment in the local playground, all offline.

Flexible Configuration

Customize project, trace, and span names. Define agents and models easily via JSON.

​
Dive Deeper

Ready to enhance your LLM development experience?

API Reference

Explore the Agensight SDK functions and classes in detail.

Practical Examples

See Agensight in action with practical use-cases and code samples.

For community support or to contribute, visit our GitHub Repository.