{"repo":"prime-radiant-inc/gauntlet","free":true,"listed":false,"github":"https://github.com/prime-radiant-inc/gauntlet","clone":"git clone https://github.com/prime-radiant-inc/gauntlet.git","description":"AI-powered QA testing framework that uses LLMs (Claude or GPT) to test web apps, CLI tools, and TUI programs from markdown story cards, returning structured pass/fail verdicts with evidence.","language":"TypeScript","stars":52,"topics":["ai-agents","browser-automation","bun","llm","qa-testing","typescript"],"license":"Apache-2.0","category":"ai-agents","readme_excerpt":"Gauntlet Gauntlet is an AI-powered QA testing framework. It uses large language models (Claude or GPT) to test software the way a human tester would: working through web apps in a real browser, CLI tools at a stdin/stdout prompt, and terminal/TUI programs in a live tmux session — clicking, typing, reading screens, and reporting bugs. You write story cards -- markdown files with a title, description, and acceptance criteria -- and Gauntlet's AI agent works through them via one of three adapters ( web , cli , tui ), delivering a verdict (pass/fail/investigate) with evidence. What it does 1. You describe what to test in a story card -- a markdown file with a title, description, and acceptance criteria. 2. An AI agent drives your target through one of three adapters: a real Chrome browser (via CDP) for web apps, a stdin/stdout subprocess for CLI tools, or a tmux-hosted session for terminal/TUI programs. 3. The agent explores and evaluates your acceptance criteria, but also reports anything else it notices: bugs, UX issues, typos, accessibility problems, performance issues, and suggestions. 4. You get a structured result with a verdict, reasoning, observations, screenshots, and an action log. Beyond single-story testing, Gauntlet can generate variations (\"fanout\") from a parent story card -- producing edge-case, error-path, and alternate-persona stories automatically. It can also generate follow-up stories from observations or failures in previous runs. Architecture Tech stack - R","default_branch":null,"files":null,"tree":[],"storefront":"/r/prime-radiant-inc","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/prime-radiant-inc/gauntlet/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}