Natural-language driven Android automation for AI coding assistants and test suites, running on real phones or emulators.
truecourse-ai/truecourse
Generates end-to-end Playwright tests from your PRDs, ADRs and READMEs, and flags when the product and its documentation disagree.
- Stars
- 538
- Forks
- 44
- Open issues
- 52
- Last commit
- 23 days ago
What is it?
TrueCourse reads the documentation a team already writes, such as PRDs, ADRs, READMEs and documentation sites, and organizes it into a set of claims. It then works out the flows a user takes through the product and writes a test for each one against the running interfaces.
When a test fails, it means the product and the documentation no longer agree, and the failure names the specific claim involved. Tests are written as Playwright specs, each with its own seed data so tests do not share state. The tool runs on a Claude Code login and includes a local mode that needs no sign-in, plus pull request checks through a GitHub App.
The project is TypeScript with a pnpm monorepo and PostgreSQL. It exposes an MCP endpoint so Claude Code can read workspace documents, flows, runs and failures.
Key features
- 01Turns PRDs, ADRs, READMEs and doc sites into a corpus of testable claims.
- 02Writes Playwright end-to-end tests for each user flow against the running product.
- 03Reports failing tests with the specific documentation claim they contradict.
- 04Checks pull requests for documentation conflicts and test failures in GitHub.
- 05Exposes an MCP endpoint so Claude Code can query workspace results.
Use cases
- Checking whether a product still behaves as its PRD and user docs describe.
- Catching drift between architecture decision records and the running system.
- Running documentation-driven checks on pull requests before merge.
Quick start
- 1
Copy the example environment file
cp .env.example .env - 2
Start Postgres with Docker Compose
docker compose up -d - 3
Install dependencies
pnpm install - 4
Start the dev server and open http://localhost:3000
pnpm dev
Things to know
- Requires a Claude Code login and the claude binary on your PATH.
- The truecourse npm CLI is deprecated and no longer maintained.
- Production use of the ee/ code needs an Enterprise subscription.
In collections
Topics
FAQ
- What is truecourse?
- Generates end-to-end Playwright tests from your PRDs, ADRs and READMEs, and flags when the product and its documentation disagree. TrueCourse reads the documentation a team already writes, such as PRDs, ADRs, READMEs and documentation sites, and organizes it into a set of claims. It then works out the flows a user takes through the product and writes a test for each one against the running interfaces.
- Is truecourse free to use? What license does it have?
- truecourse is open source under the MIT license. Check the license terms before using it in commercial products.
- Is truecourse still maintained?
- Actively maintained. The last commit to truecourse was on Oct 14, 2026.
- What can I use truecourse for?
- Checking whether a product still behaves as its PRD and user docs describe. Catching drift between architecture decision records and the running system. Running documentation-driven checks on pull requests before merge.
- How do I install truecourse?
- Copy the example environment file: cp .env.example .env Start Postgres with Docker Compose: docker compose up -d Install dependencies: pnpm install Start the dev server and open http://localhost:3000: pnpm dev
- Where can I get truecourse?
- The source code and installation instructions are on GitHub: https://github.com/truecourse-ai/truecourse
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
AI-powered end-to-end testing framework where tests are written as plain-language descriptions, built on Playwright and Anthropic Claude.
Open-source AI testing agent that turns Gherkin feature files into automated end-to-end browser, API, security, accessibility and visual tests.
Playwright is a framework for web testing and automation that drives Chromium, Firefox, and WebKit through one API.
An end-to-end testing framework for web and mobile apps that lets you describe test goals in natural language and drive the app with an agent.
A command-line interface for Playwright browser automation, designed for coding agents to drive browsers through short shell commands and skills.