Quick background on Egma

SkillMonitoring & ops

Integrate Egma into a repository containing voice agent code. Use when the developer asks to integrate voice agent testing, monitoring or both

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Quick background on Egma skill

What this skill tells your AI

The instructions your AI receives, as published by egma-ai/egma in skills/integrate-egma/SKILL.md and read by ahel’s review.

Egma is an open-source platform that helps developers and their coding agents in building reliable voice agents and making sure they work well in production. egma does it by providing two things - infrasturcture to do simulation testing to test an agent pre-release - similar to unit tests that run in CI every prod push. Other is by providing a way to monitor voice agent's behavior in production, identifying issues, etc. This skill is all about configuring the voice agents in this repo to work with egma platform.

Getting started

  1. If the egma command is unavailable, install it with npm install --global egma-cli. This is install a global egma cli. Use --help command in the cli to see all the available options.
  2. Login to the egma platform using egma login. If it is already logged in, continue. If not, use your built-in browser skills or ask the developer to authorize the login.

Setting up testing and/or monitoring based on agent platform

  • To setup testing with egma platform - refer to simulation testing setup guide
  • To setup monitoring of voice agents on egma platform - refer to monitoring setup guide
  • If the developer wants to setup both testing and monitoring - first setup monitoring and then testing - but do it as one continuous chunk of work rather than exiting. Reuse credentials across testing and monitoring unless stated otherwise.

Signals

GitHub stars
114
Forks
4
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
integrate-egma
Source
github.com/egma-ai/egma