Analysis updated 2026-05-18
Get step-by-step, screen-aware guidance while completing a task on your computer.
Use voice input and output to interact with an AI assistant hands-free.
Automate multi-step browser tasks like filling forms via the browser plugin.
Switch between AI providers (Claude, OpenAI, Gemini, Groq, Ollama) based on cost or privacy needs.
| mo-tunn/openguider | amitkumardemo/edgecareer | polymarket-sports/polymarket-trading-bot | |
|---|---|---|---|
| Stars | 144 | 142 | 142 |
| Language | JavaScript | JavaScript | JavaScript |
| Setup difficulty | easy | moderate | hard |
| Complexity | 2/5 | 3/5 | 4/5 |
| Audience | general | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Prebuilt installers are available for Windows, macOS, and Linux, an AI provider API key is needed for most features.
OpenGuider is a desktop application built with Electron that acts as an AI assistant sitting on top of whatever the user is doing on their computer. It watches the screen, can listen through voice input, and tries to guide the user step by step through real tasks by turning a stated goal into a plan and tracking progress against that plan as the situation changes. Rather than only answering questions in a chat window, it aims to give practical, action oriented help, including pointer style hints that show roughly where on screen to click next. The tool supports multiple AI providers, including Claude, OpenAI, Gemini, Groq, OpenRouter, and locally run models through Ollama, so a user can pick a provider based on speed, cost, or a preference for keeping data on their own machine. It also offers optional speech to text through services like AssemblyAI or Whisper compatible endpoints, and text to speech through Google, OpenAI, or ElevenLabs, so it can be used with voice, text, or a mix of both. A newer plugin system lets the assistant hand off certain tasks to specialized workspaces, with a browser automation plugin as the first example. That plugin can open websites, fill in forms, and carry out multi step browser tasks, either pausing for the user's approval on risky actions or running fully on its own if configured to do so. Installation is available as a prebuilt application for Windows, macOS, and Linux, downloaded from the project's releases page, or by cloning the repository and running it from source with npm. After installing, configuration happens inside the app's settings, starting with choosing a primary AI provider and entering its API key before enabling any additional features like voice or plugins. The README displays a license badge but does not spell out which specific license applies in the text itself, so anyone relying on this project for commercial use should check the actual license file in the repository before assuming particular terms apply.
An Electron desktop AI assistant that watches your screen, listens to voice, and gives step-by-step guidance for completing tasks on your computer.
Mainly JavaScript. The stack also includes Electron, JavaScript, Node.js.
License type is not stated in the README, check the repository's license file before relying on specific terms.
Setup difficulty is rated easy, with roughly 30min to a first successful run.
Mainly general.
This repo across BitVibe Labs
Verify against the repo before relying on details.