Kitsune

io.github.junyeong-nero由 GitHub 账号 junyeong-nero 发布v0.7.3更新于 Nov 5, 2026

Let your coding agent use the Firefox, Zen, or LibreWolf browser you already have open.

已验证STDIO仅桌面浏览器自动化开发者工具

概览

AI 生成的概览

让 AI 编程助手使用你已经打开的 Firefox、Zen 或 LibreWolf 浏览器,在后台标签页中操作,同时不影响你继续浏览。

可以做什么
  • 在页面上导航、点击、输入、悬停、拖拽、选择下拉项、按键和批量填写表单
  • 通过无障碍快照、文本搜索、截图和控制台消息读取页面内容
  • 查看网络请求详情并执行 JavaScript,包括运行 Playwright 代码片段
  • 管理标签页、对话框、文件上传与拖放、窗口尺寸和媒体模拟
  • 请用户登录、通过验证码或接管某一步骤,然后继续未完成的调用
适用场景
适合希望助手在你已登录的现有浏览器中做调研、填写草稿或操作网页应用,又不想安装 Playwright 或 Chromium,并且不希望自己的标签页被打扰的场景。
运行要求
通过 uvx 从 PyPI 包 kitsune-browser-mcp 在本地运行,需要安装 uv,并在 Firefox、Zen 或 LibreWolf 中安装 Kitsune 浏览器扩展。浏览器需保持打开,本地桥接会自动启动。可选环境变量 KITSUNE_MODE、KITSUNE_CAPS、KITSUNE_PUBLISH 和 KITSUNE_ASK_BEFORE_ACT 用于控制策略。未声明任何身份验证。
安装前请注意
助手运行在你已登录的浏览器配置中,因此可触及你的账号,建议使用单独的浏览器配置。browser_run_code_unsafe 会在 Playwright 服务进程中执行任意代码,被描述为等同于远程代码执行。操作可能提交表单、付款或删除内容,可用 KITSUNE_MODE 和 KITSUNE_ASK_BEFORE_ACT 加以限制。README 说明无法阻止提示注入。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Kitsune,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

Kitsune

[Kitsune: an orange fox in a browser window, connecting your browser and your AI agent]

[kitsune-browser-mcp on PyPI] [kitsune-browser on PyPI] [Python] [Release] [License: MIT] [Docs]

Let your coding agent use the browser you already have open, while you keep using it too.

Kitsune connects Claude Code, Codex, Pi, OpenCode and other MCP clients to Firefox, Zen or LibreWolf. The agent works in background tabs of its own, signed in wherever you are signed in. It can read your tabs, but it has to ask before it changes one. There is no Playwright, Chromium or WebDriver to install.

[Claude Code asks for the user's Wikipedia tab after researching in background tabs; the user approves in Kitsune's toolbar popup, the agent jumps the tab to the right section, and the user takes the tab back and closes the agent's tabs from the popup]

New in 0.7.2: the agent asks for your tab, and you approve, take it back and clean up from the toolbar popup. A real recording on Wikipedia, sped up while the agent thinks (full 45-second video).

Why Kitsune?

  • It works beside you. New tabs open in the background, so the page you are on stays in front. Your tabs, and tabs other agents opened, are read-only to the agent: it can read them, but it cannot click, type, scroll, navigate or close them. If you start typing in one of the agent's own tabs, it waits.
  • You stay in charge. When the agent needs one of your tabs, it asks, and you answer in Kitsune's toolbar popup. You can take the tab back at any time. The popup also shows whether the agent is connected, and closes all of the agent's tabs with one click. Tabs the agent leaves idle close by themselves after five minutes (you can change this in the popup).
  • The tools are familiar. The agent starts with Playwright MCP's 25 core tools, under the same names, plus browser_ask_user and browser_resume for when it needs you. Prompts written for Playwright MCP work unchanged. More tools (page text, forms, batches, downloads, …) stay off until you allow them.
  • You decide how much to hand over. KITSUNE_MODE=read, draft or act limits what the agent can do. In draft mode it can fill in forms but cannot click or submit.

Get started

  1. Install uv.

  2. Install the browser extension in Firefox, Zen or LibreWolf.

  3. Add Kitsune to your agent. For Claude Code:

    bash
    claude mcp add --scope user kitsune -- uvx --from kitsune-browser-mcp kitsune

    Codex, Pi, OpenCode, Claude Desktop and Cursor are in Installation.

  4. Keep the browser open and ask in plain language:

    Research these three products in background tabs and compare them with links. Leave my current tab alone.

    Fill in a draft of this form. Don't submit it.

Kitsune starts its local bridge on its own. If the agent cannot connect, ask it to check the connection: the error's hint says what to fix.

Use a separate browser profile for agent work and sign in only to the accounts it needs. Ask the agent to check with you before it sends, buys or deletes anything. Nothing in Kitsune stops prompt injection, so read the security guide before using sensitive accounts.

Documentation

The full documentation is at junyeong-nero.github.io/kitsune:

来源:README.md,提交 adcd511

工具

27

browser_resume

Resume a call

Continue a call whose result has call and resume_token (a needs_user stop, batch, parallel or workflow); nothing that ran ok is sent again.

输入参数call必填waitobservedresume_token必填

browser_ask_user

Ask the user

Ask the user to do what you cannot: sign in, pass a CAPTCHA (human_check), click a control that ignores your clicks (trusted_input), or choose (other).

输入参数reason必填tab_idmessage必填

browser_close

Close browser

Close the page

无需输入参数。

browser_resize

Resize browser window

Resize the browser window

输入参数width必填height必填

browser_console_messages

Get console messages

Returns all console messages

输入参数alllevel必填filename

browser_handle_dialog

Handle a dialog

Handle a dialog

输入参数accept必填promptText

browser_emulate_media

Emulate media features

Emulate CSS media features for the page, for example switch between the light and dark color scheme. Omitted parameters are left unchanged; null clears an override.

输入参数mediacontrastcolorSchemeforcedColorsreducedMotion

browser_evaluate

Evaluate JavaScript

Evaluate JavaScript expression on page or element

输入参数targetelementfilenamefunction必填

browser_file_upload

Upload files

Upload one or multiple files

输入参数paths

browser_drop

Drop files or data onto an element

Drop files or MIME-typed data onto an element, as if dragged from outside the page. At least one of "paths" or "data" must be provided.

输入参数datapathstarget必填element

browser_find

Find in page snapshot

Search the accessibility snapshot of the current page for text or a regular expression. Returns matching snapshot nodes with a few lines of surrounding context (like search snippets), each shown under its path from the root of the tree, which is cheaper than capturing the whole snapshot when you only need to locate an element and its ref.

输入参数textregexfilename

browser_fill_form

Fill form

Fill multiple form fields

输入参数fields必填

browser_press_key

Press a key

Press a key on the keyboard

输入参数key必填

browser_type

Type text

Type text into editable element

输入参数text必填slowlysubmittarget必填element

browser_navigate

Navigate to a URL

Navigate to a URL

输入参数url必填

browser_navigate_back

Go back

Go back to the previous page in the history

无需输入参数。

browser_network_requests

List network requests

Returns a numbered list of network requests since loading the page. Use browser_network_request with the number to get full details.

输入参数filterstatic必填filename

browser_network_request

Show network request details

Returns full details (headers and body) of a single network request, or a single part if `part` is set. Use the number from browser_network_requests.

输入参数partindex必填filename

browser_run_code_unsafe

Run Playwright code (unsafe)

Run a Playwright code snippet. Unsafe: executes arbitrary JavaScript in the Playwright server process and is RCE-equivalent.

输入参数codefilename

browser_take_screenshot

Take a screenshot

Take a screenshot of the current page. You can't perform actions based on the screenshot, use browser_snapshot for actions.

输入参数typescale必填targetelementfilenamefullPage

browser_snapshot

Page snapshot

Capture accessibility snapshot of the current page, this is better than screenshot

输入参数boxesdepthtargetfilename

browser_click

Click

Perform click on a web page

输入参数buttontarget必填elementmodifiersdoubleClick

browser_drag

Drag mouse

Perform drag and drop between two elements

输入参数endTarget必填endElementstartTarget必填startElement

browser_hover

Hover mouse

Hover over element on page

输入参数target必填element

browser_select_option

Select option

Select an option in a dropdown

输入参数target必填values必填element

browser_tabs

Manage tabs

Manage tabs; request a handoff.

输入参数urlindexaction必填reasontab_idconfirm_token

browser_wait_for

Wait for

Wait for text to appear or disappear or a specified time to pass

输入参数texttimetextGone

版本历史

5
  1. v0.7.3最新Nov 5, 2026
  2. v0.7.2Nov 4, 2026
  3. v0.7.1Oct 29, 2026
  4. v0.7.0Oct 25, 2026
  5. v0.4.1Oct 25, 2026

评分

已满足适用检查项的 90%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分
  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分