# Webwright Microsoft's open-source browser agent framework for long-horizon web tasks. ## Agent Decision Summary - Risk level: moderate - Source confidence: medium - Recommended workflows: Browser automation, Coding agent workflow, Evaluation and observability, Reusable skill workflow - Permission surface: browser - Agent JSON: https://www.openagent.bot/agents/webwright.agent.json ## Summary Webwright is an open-source browser agent framework from Microsoft that targets SWE-style, long-horizon web tasks. It is useful for teams comparing browser-use style agents, Playwright-based automation, and agent frameworks that need repeatable web task execution. ## Guide Webwright is Microsoft's open-source browser agent framework for long-horizon web tasks. ### What it is It is a developer framework for evaluating and building agents that operate websites across multiple steps. ### Why it matters Browser automation is one of the clearest places where agents can turn language instructions into real work, but long tasks need more than a simple click loop. ### How it works Start with the official repository, run the examples, and measure completion rate on a small set of workflows before connecting sensitive accounts. ## Use Cases - Browser workflow benchmarking: Compare how different browser agents handle multi-step web tasks. - QA automation research: Evaluate whether an agent can recover from unexpected UI states during product QA. ### FAQ - Is Webwright open source? - Yes. The GitHub repository is listed under the MIT license. - Who should evaluate Webwright? - Teams researching or prototyping long-horizon browser agents should put Webwright on the shortlist. ## What It Does It is a developer framework for evaluating and building agents that operate websites across multiple steps. ## How To Evaluate Start with the official repository, run the examples, and measure completion rate on a small set of workflows before connecting sensitive accounts. ## Why It Matters Browser agents are moving from demos into real workflows. Webwright matters because it focuses on longer tasks where planning, state, and recovery are more important than one-off page actions. ## Best For - Teams evaluating browser agents for complex web workflows - Researchers benchmarking long-horizon web task performance - Developers who want a browser automation framework with agent-oriented primitives ## Not For - Teams that only need a simple scraper - Users looking for a no-code browser automation product ## What It Actually Does - Long-horizon browser focus: Webwright is positioned around multi-step web tasks rather than isolated browser commands. - Why it matters: Long-running workflows need state tracking, retries, and task decomposition that basic automation scripts often lack. - Research-friendly surface: The project is useful for comparing browser agent performance across repeatable tasks. - Why it matters: Teams can use it as a benchmarkable alternative to ad hoc browser agent demos. - Microsoft backing: The repository is maintained under the Microsoft GitHub organization. - Why it matters: A major maintainer increases the odds of documentation, issue visibility, and ecosystem attention. ## Typical Use Cases - Web workflow evaluation: Run representative web tasks and compare completion rate, latency, and recovery behavior. - Browser agent research: Study planning and execution patterns for long web sessions. - Automation prototypes: Prototype web task agents before deciding whether to adopt a productized automation stack. ## How It Compares - Choose Webwright for long-horizon experiments vs browser-use: browser-use is a popular practical browser agent library. Webwright is especially interesting when the test workload resembles SWE-style long web tasks. ## Fit Matrix - Browser automation: strong. Webwright has multiple signals for browser automation, including matching tags, capabilities, category, or positioning. Required check: Run one non-sensitive website task and inspect clicks, waits, retries, and changed URLs. - Coding agent workflow: strong. Webwright has multiple signals for coding agent workflow, including matching tags, capabilities, category, or positioning. Required check: Run a small repository change and inspect the diff, tests, and rollback path. - Evaluation and observability: strong. Webwright has multiple signals for evaluation and observability, including matching tags, capabilities, category, or positioning. Required check: Add one repeatable test case and confirm results can run again in review or CI. - Reusable skill workflow: strong. Webwright has multiple signals for reusable skill workflow, including matching tags, capabilities, category, or positioning. Required check: Run one skill end to end and check whether it produces evidence or structured output. - Connector or protocol layer: weak. Webwright is not primarily positioned for connector or protocol layer in the current metadata. Required check: Connect one low-risk service, then inspect schemas, auth scope, errors, and logs. - Local or private AI stack: weak. Webwright is not primarily positioned for local or private ai stack in the current metadata. Required check: Verify hardware requirements, data path, storage, and whether all calls stay in your environment. ## Evidence - verified: Webwright is listed as open source. Source: License metadata: MIT - verified: Webwright has a recorded GitHub repository: microsoft/Webwright. Source: Resource facts and GitHub source link. - inferred: Webwright is tagged with workflow orchestration capabilities. Source: OpenAgent capability taxonomy. ## Missing Checks - Dedicated docs link is missing. - Repository freshness has not been recorded. ## Next Actions - Inspect repository: https://github.com/microsoft/Webwright ## Facts - Category: agents - Resource type: agent - Open source: yes - License: MIT - Last verified: 2026-06-09 - GitHub repo: microsoft/Webwright - GitHub stars: 5239 ## Capabilities - workflow-orchestration ## Structured Use Case Tags - developer-workflow ## Getting Started - Open the GitHub repository: https://github.com/microsoft/Webwright ## Links - GitHub: https://github.com/microsoft/Webwright ## Structured Outputs - JSON: https://www.openagent.bot/agents/webwright.json - Markdown: https://www.openagent.bot/agents/webwright.md - Agent JSON: https://www.openagent.bot/agents/webwright.agent.json - Canonical: https://www.openagent.bot/agents/webwright