playwright-interpreter 0.1.0

Interactive REPL for Playwright browser automation with automatic script generation

Playwright Interpreter

Interactive REPL for Playwright browser automation with automatic Python script generation.

Record commands as you explore a site, then export a standalone Playwright script you can run, share, or check into version control.

Installation

pip install playwright-interpreter
playwright install

For development:

git clone https://github.com/playwright-interpreter/playwright-interpreter.git
cd playwright-interpreter
pip install -e ".[dev]"
playwright install

Quick Start

Launch the interactive REPL:

playwright-interpreter

Example session:

Playwright Interpreter v0.1.0

pi> launch
pi> goto https://google.com
pi> fill textarea "playwright interpreter"
pi> click input[type=submit]
pi> screenshot result.png
pi> code
pi> exit

The code command displays the generated script:

from playwright.sync_api import sync_playwright

with sync_playwright() as p:
    browser = p.chromium.launch(headless=False)

    page = browser.new_page()

    page.goto("https://google.com")
    page.fill("textarea", "playwright interpreter")
    page.click("input[type=submit]")
    page.screenshot(path="result.png")

    browser.close()

On exit, the session is saved automatically:

• session.json — recorded commands
• generated_script.py — executable Playwright script

REPL Usage

Meta Commands

Command	Description
help	Show all available commands
history	Show executed commands
code	Display generated Python script
save	Save session.json and generated_script.py
clear	Clear recorded actions
export <file>	Export script to a custom filename
load <session.json>	Load a previous session
exit	Exit, save session, and close browser

Playwright Commands

Command	Description
launch	Launch Chromium browser (opens a new page)
new_page	Open a new browser page
goto <url>	Navigate to a URL
click <selector>	Click an element
dblclick <selector>	Double-click an element
hover <selector>	Hover over an element
fill <selector> <text>	Fill an input field
type <selector> <text>	Type text into an element
press <selector> <key>	Press a key on an element
wait <milliseconds>	Wait for a duration
reload	Reload the current page
back	Navigate back
forward	Navigate forward
screenshot <file>	Save a screenshot
close	Close the browser

Quoted strings are supported for selectors and text with spaces:

pi> fill "#search" "playwright interpreter"

CLI Commands

Interactive REPL

playwright-interpreter

Replay a Session

Re-execute recorded actions from a saved session:

playwright-interpreter replay session.json

Generate Script

Create a Playwright script from a session file without opening a browser:

playwright-interpreter generate session.json
playwright-interpreter generate session.json -o my_script.py

Examples

Save During a Session

pi> launch
pi> goto https://example.com
pi> screenshot home.png
pi> save

Load a Previous Session

pi> load session.json
pi> code

Export to Custom File

pi> export my_automation.py

Development

Run tests with coverage:

pytest

Run tests without coverage threshold:

pytest --no-cov

Project structure:

playwright_interpreter/
├── cli.py          # Typer CLI entrypoint
├── repl.py         # Interactive REPL
├── parser.py       # shlex-based command parser
├── commands.py     # Command definitions
├── executor.py     # Playwright command execution
├── session.py      # Browser session management
├── recorder.py     # Action recording and persistence
├── generator.py    # Python script generation
├── models.py       # Pydantic data models
├── exceptions.py   # Custom exceptions
└── utils.py        # Shared utilities

Publishing

Build and publish to PyPI:

pip install build twine
python -m build
twine upload dist/*

Ensure Playwright browsers are installed on target machines:

playwright install

License

MIT License — see LICENSE for details.