firefox-devtools-mcp
Related Servers
Alternatives to firefox-devtools-mcp
- AlicenseCqualityDmaintenanceProvides Firefox browser automation capabilities through Playwright, allowing control of Firefox from MCP-enabled applications with features for basic automation and advanced multi-tab debugging.2929MIT
Related Servers
- FlicenseAqualityDmaintenanceModel Context Protocol server for automating Firefox via WebDriver BiDi, enabling page navigation, snapshots, input, network monitoring, and console debugging for AI agents in MCP clients.315,487 npm-
- AlicenseBqualityBmaintenanceEnables AI agents to drive a user-launched Firefox instance via its native Marionette protocol, providing precise DOM actuation like clicking, typing, uploading files, taking screenshots, and evaluating JavaScript.1813 npm1GPL 2.0
- AlicenseAqualityCmaintenanceDrive Firefox-based browsers (Floorp, LibreWolf, Zen, Waterfox, Mullvad, Firefox) from any MCP client — read pages, screenshot, click, fill forms and manage tabs in your real session, over Marionette/WebDriver. OS input & JS eval locked by default.4168 npm2MIT
- AlicenseNot gradedqualityBmaintenanceEnables an MCP client to drive a real, already-logged-in Firefox session through a WebExtension rather than a WebDriver connection, so tabs, cookies and logins work as-is without relaunching the browser under remote-control flags. Exposes tools for tab listing and selection, navigation, accessibility-style snapshots, clicking, filling, typing, key presses, file uploads, screenshots and waiting, with per-origin permission grants and opt-in script evaluation.MIT
- AlicenseBqualityAmaintenanceEnables AI agents to automate and control a real Firefox instance over WebDriver BiDi through 177 browser_* tools spanning web pages, tabs, windows, bookmarks and history, settings, extensions, DevTools, translation, certificates, and more across 17 domains.178MIT
- AlicenseBqualityCmaintenanceLets Claude Code attach to a running, signed-in Firefox over WebDriver BiDi to control tabs, navigate, screenshot, read pages, click, type, fill forms, evaluate JavaScript, and inspect console/network activity.22MIT
TDQS
Scored across 35 tools
Most tools target a clearly distinct resource+action (tabs, snapshots, screenshots, dialogs, downloads, extensions), and the descriptions explicitly disambiguate the risky pairs (fill_by_uid vs type_text vs press_key, get_page_text vs evaluate_script vs take_snapshot). The remaining overlaps are mild: three different read/extraction paths and two screenshot variants, plus clear_snapshot/resolve_uid_to_selector which are niche UID helpers an agent may rarely reach for.
The dominant pattern is consistent snake_case verb_noun (get_page_text, list_pages, close_page, set_viewport_size, click_by_uid). Deviations are minor but real: screencast_start/screencast_stop use noun_verb ordering, and screenshot_page/print_page treat the noun as the verb, while new_page uses an adjective-first form.
At 35 tools this is heavy relative to the 3-15 sweet spot, even though the browser-automation domain genuinely spans tabs, interaction, capture, downloads, extensions, and diagnostics. A few tools feel like marginal additions (clear_snapshot, resolve_uid_to_selector, get_firefox_info) that could be folded into snapshot/info tooling to slim the surface.
Coverage is broad and coherent: navigation and history, tab lifecycle, DOM snapshot plus UID-driven interaction, screenshots/PDF, downloads, dialogs, extensions, viewport, screencasts, JS evaluation, and Firefox log output. Gaps are minor rather than blocking: no cookie/localStorage management, no network request inspection or interception, and no explicit accessibility-tree or console-message retrieval beyond raw Firefox output.