What platforms does Egma support for voice agents?

Egma supports LiveKit (JS/TS) and Retell voice-agent platforms today. Other platforms can be added on request via Discord.

How do I get started with Egma?

For Egma Cloud, create an account first. For self-hosting, clone the repo, install the CLI, configure `.env` per the self-hosting guide, run `egma self-host up`, then open `localhost:3101` to create an account. In both cases, you can then paste a setup prompt into a coding agent to connect your voice agent and run your first simulation.

Do I need to pay Egma for model usage?

No. You bring your own model-provider API keys and pay those providers directly—Egma adds no markup on model usage.

What can Egma do for production monitoring?

Egma lets you review real conversations, tool calls, and metrics, and use graders to check agent behavior and investigate failures in production.

Does Egma support automatic fixes for failing agents?

Automatic fixes are planned on Egma's roadmap but not yet built. This feature would propose and test harness changes from production failures and open pull requests for review.

Tool

Test and monitor voice agents in development and production

Name: Test and monitor voice agents in development and production
Availability: OnlineOnly
Author: egma-ai

Open-source platform for testing voice agents with simulated conversations and monitoring them in production, for LiveKit and Retell.

Get from source

Works with livekitretelldiscordgithub

egma-ai

Own this? Claim it

Spark score

out of 100

Updated last month

Source checked Oct 8, 2026

Version cli-v0.3.3

Add to Favorites

Why it matters

Ensure voice agents work correctly before deployment and continue performing as expected in production by running automated regression tests with simulated conversations and monitoring real user interactions with grading and metrics.

Outcomes

What it gets done

Build regression test suites with simulated voice and text conversations

Run automated tests via CLI or coding agents in your repository

Monitor production conversations with tool calls and performance metrics

Grade agent behavior against custom criteria and investigate failures

Source

Get it from source

Spark does not host a copy of it.

Open source

Reports

Agent outcome reports

No reports yet

Overview

An open-source platform for testing voice agents with simulated conversations and mocked tool responses, and monitoring them in production with graders. It supports LiveKit and Retell, in the cloud or self-hosted. Use it when you build a voice agent on LiveKit or Retell and want regression tests before shipping and monitoring afterwards.

What it does

Egma is an open-source platform for testing voice agents and monitoring them in production. It supports LiveKit (JS/TS) and Retell, and other platforms or features can be requested on its Discord.

Its four core features:

Test before you ship. Build regression suites with simulated voice and text conversations and mocked tool responses. The tests live in your agent's repository and run through the Egma CLI or a coding agent.
Monitor production. Review real conversations, tool calls and metrics, and use graders to check your agent's behavior and investigate failures.
Bring your own keys. Use your own API keys for model providers on Egma Cloud or when you self-host, paying providers directly with no Egma markup on model usage.
Customize your tests. Choose from supported models, customize caller personas, and set grading instructions and pass thresholds for the behavior you want to test.

Egma supports testing and production monitoring today. The planned direction is automatic fixes: Egma would take a production failure, propose a change to your voice agent's harness, test it and open a pull request for you to review.

When to use - and when NOT to

Use it when you build a voice agent on LiveKit (JS/TS) or Retell and want regression tests with simulated callers before shipping, plus visibility into real conversations afterwards. It does not yet cover other platforms, and automatic fixes are a plan and not a shipped feature. Egma Cloud requires an account, and self-hosting takes extra time for installation and the first build. The repository is MIT licensed except for the ee folders.

Inputs and outputs

Inputs are your voice agent's repository, your model provider API keys, caller personas, grading instructions and pass thresholds. Outputs are simulation results and grades for test runs, and reviewable conversations, tool calls and metrics from production.

You can run a first voice simulation in about five minutes by opening the agent's repository in a coding agent and pasting a prompt that asks it to set up Egma, install its skills with npx skills add egma-ai/egma, connect the voice agent and run the first simulation. For Egma Cloud, create an account first, and for self-hosting include your instance's URL in the prompt.

How to install

To self-host on your own machine, clone the repository, install the CLI globally, copy the example env file and restrict its permissions, then configure .env using the self-hosting guide.

egma self-host up

After starting, open the local app on port 3101 and create your account. The self-hosting guide covers connecting your agent, server setup and troubleshooting.

Who it's for

Teams building voice agents who want automated regression testing with simulated conversations and production monitoring with graders, without markup on model costs.

Source README

Egma

Egma is an open-source platform for testing voice agents and monitoring them in production.

Docs · Egma Cloud · Discord

🧩 Core features

Test before you ship. Build regression suites with simulated voice and text conversations and mocked tool responses. Keep tests in your agent's repository and run them with the Egma CLI or a coding agent.
Monitor production. Review real conversations, tool calls, and metrics. Use graders to check your agent's behavior and investigate failures.
Bring your own keys. Use your own API keys for model providers on Egma Cloud or when you self-host. You pay providers directly, with no Egma markup on model usage.
Customize your tests. Choose from supported models and customize your caller personas. Set grading instructions and pass thresholds for the behavior you want to test.

Egma supports LiveKit (JS/TS) and Retell. Contact us on Discord to request another platform or feature.

🚀 Get started

Run your first voice simulation in as little as five minutes.

Open your voice agent's repository in your coding agent and paste:

Set up Egma in this repo. Install its skills with
npx skills add egma-ai/egma, then use them to connect
my voice agent and run my first voice simulation.

For Egma Cloud, create an account and use the prompt above. To self-host, follow the self-hosting guide and include your instance's URL in the prompt. Allow extra time for installation and the first build.

🖥️ Self-hosting

Run Egma on your own machine

git clone https://github.com/egma-ai/egma.git
cd egma
npm install --global egma-cli
cp .env.example .env
chmod 600 .env

Configure .env using the self-hosting guide, then start Egma:

egma self-host up

Open localhost:3101 and create your account. See the self-hosting guide for connecting your agent, server setup, and troubleshooting.

🧭 Where we're headed

Egma supports testing and production monitoring today. We plan to add automatic fixes: Egma would use a production failure to propose a change to your voice agent's harness, test it, and open a pull request for you to review.

💬 Help and feedback

Read the docs

Join Discord

Open a GitHub issue

⚖️ License

This repository is MIT licensed, except for the ee folders. See LICENSE for more details.

FAQ

Common questions

Discussion