{"repo":"lightpanda-io/browser","free":true,"listed":false,"github":"https://github.com/lightpanda-io/browser","clone":"git clone https://github.com/lightpanda-io/browser.git","description":"Lightpanda: the headless browser designed for AI and automation","language":"Zig","stars":33937,"topics":["browser","browser-automation","cdp","headless","lightpanda","playwright","puppeteer","zig"],"license":"AGPL-3.0","category":"web_scraping_engine","readme_excerpt":"

\n \"Logo\"\n

\n

Lightpanda Browser

\n

\nThe headless browser built from scratch for AI agents and automation.
\nNot a Chromium fork. Not a WebKit patch. A new browser, written in Zig.\n

\n\n\n
\n\n[![License](https://img.shields.io/github/license/lightpanda-io/browser)](https://github.com/lightpanda-io/browser/blob/main/LICENSE)\n[![Twitter Follow](https://img.shields.io/twitter/follow/lightpanda_io)](https://twitter.com/lightpanda_io)\n[![GitHub stars](https://img.shields.io/github/stars/lightpanda-io/browser)](https://github.com/lightpanda-io/browser)\n[![Discord](https://img.shields.io/discord/1391984864894521354?style=flat-square&label=discord)](https://discord.gg/K63XeymfB5)\n\n
\n
\n\n[\n](https://github.com/lightpanda-io/demo)\n \n[\n](https://github.com/lightpanda-io/demo)\n
\n\n## Benchmarks\n\nRequesting 933 real web pages over the network on a AWS EC2 m5.large instance.\nSee [benchmark details](https://github.com/lightpanda-io/demo/blob/main/BENCHMARKS.md#crawler-benchmark).\n\n| Metric | Lightpanda | Headless Chrome | Difference |\n| :---- | :---- | :---- | :---- |\n| Memory (peak, 100 pages) | 123MB | 2GB | ~16 less |\n| Execution time (100 pages) | 5s | 46s | ~9x faster |\n\n## Quick start\n\n### Install\n\n**Package Managers**\n\nLatest nightly from Homebrew:\n```console\nbrew install lightpanda-io/browser/lightpanda\n```\n\nLatest nightly from Arch Linux User Repository:\n```console\nyay -S lightpanda-nightly-bin\n```\n\n**Download from the nightly builds**\n\nYou can download the last binary from the [nightly\nbuilds](https://github.com/lightpanda-io/browser/releases/tag/nightly) for\nLinux and MacOS for both x86_64 and aarch64.\n\n*For Linux*\n```console\ncurl -L -o lightpanda https://github.com/lightpanda-io/browser/releases/download/nightly/lightpanda-x86_64-linux && \\\nchmod a+x ./lightpanda\n```\n\nVerify the binary before running anything:\n```console\n./lightpanda version\n```\n\n[Linux aarch64 is also available](https://github.com/lightpanda-io/browser/releases/tag/nightly)\n\n> **Note:** The Linux release binaries are linked against glibc. On musl-based distros (Alpine, etc.) the binary fails with `cannot execute: required file not found` because the glibc dynamic linker is missing. Use a glibc-based base image (e.g., `FROM debian:bookworm-slim` or `FROM ubuntu:24.04`) or [build from sources](#build-from-sources).\n\n*For MacOS*\n```console\ncurl -L -o lightpanda https://github.com/lightpanda-io/browser/releases/download/nightly/lightpanda-aarch64-macos && \\\nchmod a+x ./lightpanda\n```\n\n[MacOS x86_64 is also available](https://github.com/lightpanda-io/browser/releases/tag/nightly)\n\n*For Windows + WSL2*\n\nLightpanda has no native Windows binary. Install it inside WSL following the Linux steps above.\n\nWSL not installed? Run `wsl --install` from an administrator shell, restart, then open `wsl`.\nSee [Microsoft's WSL install guide](https://learn.microsoft.com/en-us/windows/wsl/install) for details.\n\nYour automation client (Puppeteer, Playwright, etc.) can run either inside WSL or on the Windows host. WSL forwards `localhost:9222` automatically.\n\n**Install from Docker**\n\nLightpanda provides [official Docker\nimages](https://hub.docker.com/r/lightpanda/browser) for both Linux amd64 and\narm64 architectures.\nThe following command fetches the Docker image and starts a new container exposing Lightpanda's CDP server on port `9222`.\n```console\ndocker run -d --name lightpanda -p 127.0.0.1:9222:9222 lightpanda/browser:nightly\n```\n\n### Dump a URL\n\n```console\n./lightpanda fetch --obey-robots --dump html --log-format pretty --log-level info https://demo-browser.lightpanda.io/campfire-commerce/\n```\n\nYou can use `--dump markdown` to convert directly into markdown.\n`--wait-until`, `--wait-ms`, `--wait-selector` and `--wait-script` are\navailable to adjust waiting time before dump.\n\n### Start a CDP server\n\n```console\n./lightpanda serve --obey-robots --log-format pretty --log-level info --host 127.0.0.1 --port 9222\n```\nOnce the CDP server started, you can run a Puppeteer script by configuring the\n`browserWSEndpoint`.\n\n
\nExample Puppeteer script\n\n```js\nimport puppeteer from 'puppeteer-core';\n\n// use browserWSEndpoint to pass the Lightpanda's CDP server address.\nconst browser = await puppeteer.connect({\n browserWSEndpoint: \"ws://127.0.0.1:9222\",\n});\n\n// The rest of your script remains the same.\nconst context = await browser.createBrowserContext();\nconst frame = await context.newPage();\n\n// Dump all the links from the frame.\nawait frame.goto('https://demo-browser.lightpanda.io/amiibo/', {waitUntil: \"networkidle0\"});\n\nconst links = await frame.evaluate(() => {\n return Array.from(document.querySelectorAll('a')).map(row => {\n return row.getAttribute('href');\n });\n});\n\nconsole.log(links);\n\nawait frame.close();\nawait context.close();\nawait browser.disconnect();\n```\n
\n\n### Agent mode\n\n`lightpanda agent` lets you drive the browser with a native agent. Describe what\nyou want in plain English or with slash commands, and it controls the browser:\nnavigating pages, clicking through flows, filling forms, extracting structured\ndata. Think of it as a robot you're directing to use the web, more than a\nchatbot you're having a conversation with.\n\nBecause the agent runs inside the same process as the browser, every tool call\nis a direct operation and you retain Lightpanda's speed and memory advantage.\n\nThe output of an agent session is a\n[PandaScript](https://lightpanda.io/docs/usage/pandascript): vanilla JavaScript\nwith a small set of native browser primitives built directly into Lightpanda.\nRun `/save` to export one from your current session, then replay it with\n`lightpanda run