# microsoft/Webwright

A simple SWE style browser agent framework that achieves SOTA results on long horizon web tasks.

Repository: https://github.com/microsoft/Webwright
Canonical: https://ross.abutalabs.com/products/webwright
Language: Python
License: MIT
License Family: permissive
Last push: 2026-08-03T22:00:27+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 95, release rhythm 35, longevity 10
- inputs: {"age_days": 148, "days_push": 30, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5947, forks 380 (observed 2026-08-28T04:09:32.688172+00:00)

## What it is
Webwright is a Python framework from Microsoft that turns coding LLMs into state-of-the-art browser agents by giving them a terminal to launch and control Playwright browser sessions. Each web task is completed end-to-end as a single re-runnable Python script, with optional skill reuse and integrations for Claude Code, Codex, and other agent harnesses.

## Use cases
- automate long-horizon web tasks with an llm browser agent
- turn coding models like claude code into browser agents
- run web tasks end-to-end as re-runnable python scripts
- reuse solved web tasks as standalone code skills without tokens
- benchmark web agents on webarena-style tasks
- control multiple browser sessions from a terminal with playwright

## When to choose
- you want a minimal, script-based browser agent without multi-agent orchestration
- you need state-of-the-art results on long-horizon web tasks
- you want to extend existing coding agents with browser capabilities
- you prefer reusable, verified code skills over repeated model calls

## When to avoid
- you need a no-code visual browser automation tool
- you only need simple scraping without an LLM
- you require a mature GUI or managed cloud service

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, web-scraping, llm-inference, workflow-automation, browser-extension
- domain: large-language-models, web-development
- platform: python, cross-platform, cli
- tags: browser-agent, playwright, web-automation, swe-agent, claude-code-plugin, skill-factory, ai-agents, automation

## Member repositories
- microsoft/Webwright (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:32.688172+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:50:44.045743+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/Webwright (fetched 2026-08-28T04:09:32.688172+00:00, sha f98d20d22d99)
- Data as of 2026-08-30T08:39:29.467469+00:00.