/jump to repo

Repos Recent Bookmarks Watched Notes Tags Discover Compare Stats GitHub

Repository brief

microsoft/playwright-mcp

Read the upstream summary on the left, browse the cached forks below it, and load each fork comparison into the right-hand panel.

Cached analysis

cached 2026-03-30T20:04:23.012Z

1mo ago

microsoft/playwright-mcp

`microsoft/playwright-mcp` is a Node.js Playwright MCP server for browser automation. It is active, widely used, and maintained in the open, with 30,003 stars, 2,417 forks, and recent updates as of 2026-03-30. The repo appears suited to forks that want to extend or specialize MCP-based browser automation rather than rebuild the core server from scratch.

Loading tags...

Stars30,003

Forks2,417

Default branchmain

Last pushed2026-03-28T01:05:10Z

Recommended shortcuts

Jump straight into Discofork's strongest cached fork picks, or open a compare view in one click.

Forks

Choose a fork to inspect

10 of 10 fork briefs

Maintenance:

Magnitude:

Sort:

Selected

Choose this fork if you want a cloud-hosted browser MCP workflow and are comfortable with significant divergence from upstream. Choose upstream if you want the current, broadly maintained Playwright MCP server with the standard local-first setup and lower integration risk.

Prefer this fork if anti-bot bypass and an HTTP/OpenAPI interface are core requirements. Prefer upstream if you want the latest maintained Playwright MCP behavior, extension work, and lower operational risk.

Choose this fork if your main problem is multi-project browser automation with isolated state and persistent auth. Choose upstream if you want the most current, simplest, and best-maintained Playwright MCP baseline.

Choose this fork if you want broader browser-control and extraction features than upstream and can tolerate drift. Choose upstream if you want the actively maintained, safer default with recent fixes and packaging updates.

Choose this fork if you want a customized browser automation platform with extension, cloud, and research workflows built in. Choose upstream if you want the smallest, most current, easiest-to-track Playwright MCP server.

Choose this fork if you need customized Playwright MCP behavior, especially around extension-driven workflows and browser reuse. Choose upstream if you want the newest fixes, packaging, and the least maintenance risk.

Choose this fork if you want extra browser-automation features and are comfortable with significant divergence from upstream. Choose upstream if you want the most current, best-supported, lowest-risk Playwright MCP server.

Choose this fork if you value added extensibility and custom workflows more than staying close to upstream. Choose upstream if you want the safer path for compatibility, freshness, and lower maintenance burden.

Choose this fork if Electron support is the requirement and you want a package that installs more like a prebuilt distribution. Prefer upstream if you need the latest maintenance, broader client docs, and lower long-term upgrade risk.

Choose upstream unless you specifically need a static snapshot. This fork adds no visible capability and is materially behind, so it is a poor default for new adopters.

Fork comparison

browsermcp-com/mcp

38/100

stale

significant_divergence

Choose this fork if you want a cloud-hosted browser MCP workflow and are comfortable with significant divergence from upstream. Choose upstream if you want the current, broadly maintained Playwright MCP server with the standard local-first setup and lower integration risk.

Likely purpose

A cloud-hosted browser MCP fork that shifts the upstream Playwright MCP server toward browsermcp.com’s hosted/browser-connection workflow, with substantial extension and relay changes for a different operational model.

Best for

Teams that want a hosted browser automation backend instead of running the upstream server locally.; Adopters building browser workflows around a proprietary or managed browser service.; Users who value a custom browser connection/extension pipeline over upstream parity.; Organizations willing to trade upstream simplicity for a more productized cloud MCP experience.

Additional features

Cloud-hosted browser MCP workflow tied to browsermcp.com.
Browser extension relay and connection management changes in `src/extension/cdpRelay.ts` and `src/tab.ts`.
Extension UI and selector-related work, including page selection and connect flow changes.
Operational tooling additions such as eval/script and session-log support.
Devcontainer and packaging updates that make the fork easier to run in a managed environment.
Adds plugin, tool, or MCP-style extensibility beyond the upstream baseline.

Missing features

Materially lags upstream with 200 commits behind and 200 ahead, so adopters will not get current upstream fixes and releases as-is.
Upstream-focused README and packaging/docs are heavily rewritten, which may make standard Playwright MCP setup and guidance harder to follow.
Upstream extension tests and fixture coverage appear removed or replaced in the fork diff, reducing confidence in parity for the original local-extension workflow.
The fork appears to trade the upstream lightweight local-first setup for a cloud/browser-service oriented path, so users wanting the stock self-hosted MCP server may find it less direct to use.
It trails upstream by 200 commits, so some recent upstream features and fixes are likely not present yet.

Strengths

Likely better fit for hosted browser operations than the stock local MCP server.
Substantial work around browser connection and extension behavior suggests a more opinionated end-to-end workflow.
Recent enough to indicate active fork work, but not synchronized with upstream.
Could be attractive if you want a browser service wrapper rather than the upstream reference implementation.

Risks

Large bidirectional divergence makes upstream merges and security updates harder.
The fork is far behind upstream in both directions, so behavior may be inconsistent with current Playwright MCP expectations.
If your automation depends on upstream local/server semantics, this fork may introduce hidden compatibility gaps.
Operational dependence on a cloud service can add vendor lock-in and availability risk.