/jump to repo

Repos Recent Bookmarks Watched Notes Tags Discover Compare Stats GitHub

Repository brief

ChromeDevTools/chrome-devtools-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 2026-03-30T15:52:21.967Z

1mo ago

ChromeDevTools/chrome-devtools-mcp

chrome-devtools-mcp is a Node.js/TypeScript MCP server that lets coding agents control and inspect a live Chrome browser. It focuses on browser automation, debugging, and performance analysis, and ships as an npm package with active recent development, large adoption, and broad fork activity.

Loading tags...

Stars32,371

Forks1,914

Default branchmain

Last pushed2026-03-30T15:12:54Z

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

Prefer this fork only if you need its extra CLI/runtime hooks or its older customized behavior. If you want the latest Chrome DevTools MCP feature set, especially recent page/tool listing fixes and upstream maintenance, upstream is the safer choice.

Prefer this fork if you want a customized agent-bridge workflow with longer waits, cross-discussion, and extra orchestration features. Prefer upstream if you want the latest Chrome DevTools MCP fixes, better compatibility, and lower maintenance risk.

Choose this fork only if its custom workflow or simplification is the point. For most adopters, upstream is the safer default because this fork is significantly behind and appears to have trimmed major bundled tooling.

Choose this fork only if you want the extra agent/workflow scaffolding and are willing to own a large merge and maintenance burden. If you mainly want Chrome DevTools MCP itself, upstream looks safer and more current.

Prefer this fork only if you need its older compatibility or onboarding tweaks. For most adopters, upstream is the better choice because it is far more current and has recent feature and bugfix work this fork lacks.

Choose this fork if Brave support is the main requirement. Choose upstream if you want the latest MCP behavior, faster fixes, and lower maintenance risk.

Choose upstream unless you specifically need this fork's namespace or a frozen copy. This fork shows no added capabilities and is behind current upstream, so adopters mainly inherit the same product with recent improvements missing.

Choose this fork if your priority is remote-browser deployment and custom MCP configuration. Choose upstream if you want the newest fixes, lower maintenance burden, and the most validated performance/debugging path.

Choose this fork if you want a customized MCP server with extra in-page tool visibility and are comfortable owning compatibility and update risk. Choose upstream if you want the latest fixes, tighter browser support, and lower maintenance overhead.

Choose this fork only if stealth or specialized browser workflows matter more than staying aligned with upstream. For most adopters who want a maintained Chrome DevTools MCP server with the latest fixes, upstream is the safer choice.

Fork comparison

K3100690pdi/devtools-mcp

38/100

stale

significant_divergence

Prefer this fork only if you need its extra CLI/runtime hooks or its older customized behavior. If you want the latest Chrome DevTools MCP feature set, especially recent page/tool listing fixes and upstream maintenance, upstream is the safer choice.

Likely purpose

To customize Chrome DevTools MCP for a narrower or older operational profile, with added CLI/runtime conveniences and some tool behavior changes, while diverging substantially from upstream’s current feature set.

Best for

Developers maintaining a customized MCP server for internal agent workflows.; Users who need Deno or Bun compatibility.; Teams that value CLI/debugging tweaks over staying on the upstream release train.; Adopters willing to trade upstream parity for a more specialized fork.

Additional features

Network request filtering by resource type.
A `--debug-file` CLI option.
Compatibility work to unblock use with Deno and Bun.
More detailed performance trace summaries, including data format information.
Performance/trace descriptions that mention core web vitals.
Improved MCP client setup guidance for Gemini CLI.

Missing features

It is 200 commits behind upstream, so it misses recent upstream fixes and additions such as in-page tool listing responses for page actions.
It likely lacks upstream fixes around `.mcp.json` configuration handling and `list_pages` after closing a selected page.
The fork deletes the bundled Lighthouse devtools MCP bundle and related notices, so upstream performance-analysis packaging appears removed or substantially changed.
It is far behind upstream’s current maintenance stream, so adopters should expect missing recent bug fixes and docs updates.
It trails upstream by 200 commits, so some recent upstream features and fixes are likely not present yet.

Strengths

Adds practical CLI and runtime options that make local experimentation easier.
Includes network filtering capabilities that can help reduce noise during debugging.
Shows attention to trace output quality and debugging ergonomics.
May appeal to users who want broader runtime compatibility than upstream’s default Node-centric path.

Risks

Large divergence from upstream with 220 changed files and a big deletion footprint.
Performance-analysis behavior may differ materially because Lighthouse bundling was removed.
Being 200 commits behind upstream raises compatibility and maintenance risk.
Some upstream fixes around MCP response formatting and page lifecycle may be absent.