What is ai-qa-orchestrator?

ai-qa-orchestrator is a Python based automation pipeline that turns written user stories into runnable software tests without a developer having to write them manually. A user story is a short description of a feature from the end user's perspective, and acceptance criteria are the specific conditions that mark a story as done. The pipeline feeds a user story to Claude AI, which reads the story and automatically generates multiple categories of test cases: the happy path, meaning the normal expected flow, edge cases, meaning unusual but valid inputs, and negative test cases, meaning inputs that should fail gracefully. It then writes those as a ready to run Playwright TypeScript spec file. Playwright is a browser automation tool used to simulate real user interactions with a web application. The architecture described in the README has three layers: an MCP tools layer connecting to external services including Jira, GitHub, TestRail, and Slack, an AI orchestrator layer where Claude reasons and writes the tests, and a planned RAG knowledge layer that would let the AI reference your own codebase and coding standards. MCP, or Model Context Protocol, and RAG, or Retrieval Augmented Generation, are methods for giving AI models access to external data at runtime. To run it, the README shows installing the anthropic Python package, setting an API key, and running orchestrator.py. The output is a TypeScript spec file named after the story identifier. The Jira connector, automatic GitHub PR filing, and RAG layer are listed on the roadmap as not yet complete. The license is MIT.