Skip to content

Visualizing Simulations

The Simulations visualizer in LangWatch provides a powerful way to inspect and analyze the results of your agent tests built with the scenario library. It offers a user-friendly interface to dig into simulation runs, helping you debug your agent’s behavior and collaborate with your team.

Why Use the Simulations visualizer?

  • See every conversation between your agents and simulated users
  • Debug tool calls and agent decisions step-by-step
  • Browse, filter, and search through all your simulation runs
  • Share and review results with your team
  • Spot issues and improve your agents faster

How to Get Started

  1. Get your API key: Setup page

    Tip: Add your LANGWATCH_API_KEY to your environment right away so your results are visualized automatically. Get your key here.

  2. Set LANGWATCH_API_KEY in your environment variables
  3. Run your scenario tests as usual
  4. Open the Simulations Visualizer: https://app.langwatch.ai/@project/simulations

Simulation Results (Detailed View)

Simulation Results

  • Inspect every message, tool call, and judge evaluation
  • See exactly why a scenario passed or failed

Simulation Set Overview (Dashboard)

Simulation Set Overview

  • Get a bird's-eye view of all your scenario runs
  • Quickly spot patterns, successes, and failures

Help & Tips

Tips & Best Practices

  • Always set your LANGWATCH_API_KEY to enable reporting
  • Use the visualizer after every major change to your agent
  • Share links to specific runs with your team for review

FAQ

Q: Do I need a paid plan?
A: The Simulations Visualizer is available to all users!

Q: Where do I get my API key?
A: Get your key here or from the setup screen after signup.

Q: Can I use the visualizer for both Python and JS scenarios?
A: Yes! As long as you set your LANGWATCH_API_KEY, all results are reported and visualized.

Ready to explore?

👉 Open the Simulations Visualizer