Tabbit开发工具

tabbit-devtools

by carri1sun

Use Tabbit with agent-browser by reading Tabbit's live DevToolsActivePort file, deriving the browser wsEndpoint, and routing browser actions through agent-browser --cdp.

4.2k效率与工作流未扫描2026年4月6日

安装

claude skill add --url https://github.com/openclaw/skills

文档

Tabbit Devtools

Prefer this skill whenever the request is explicitly about Tabbit or includes phrases like 用我的 tabbit 浏览器, 在 Tabbit 里, Tabbit 当前页, or Tabbit 当前标签.

Treat Tabbit as a Chromium-based browser. This skill is about how to connect agent-browser to Tabbit. After the connection is established, handle browser automation and inspection through the normal agent-browser workflow. Do not implement a parallel browser automation layer, bridge daemon, or custom CDP client inside this skill.

agent-browser Quick Reference

Treat agent-browser as the browser-operation layer after Tabbit endpoint discovery.

The most relevant commands for this skill are:

  • open <url>
  • snapshot -i
  • click @e3
  • fill @e5 <text>
  • press Enter

Do not restate a full agent-browser manual here. Use these commands as the default vocabulary for Tabbit tasks, and prefer the official README for any broader command surface.

Quick Path

  1. Read ~/Library/Application Support/Tabbit/DevToolsActivePort first.
  2. If that file does not exist, read ~/Library/Application Support/Tabbit Browser/DevToolsActivePort.
  3. Use both lines in that file:
    • line 1: TCP port
    • line 2: browser path such as /devtools/browser/<id>
  4. Build the full browser endpoint as ws://127.0.0.1:<port><path>.
  5. Prefer that wsEndpoint over http://127.0.0.1:<port>. Tabbit may expose the browser WebSocket while /json/version and /json/list still return 404.
  6. Prefer scripts/run_agent_browser_on_tabbit.py for actual browser actions. It injects the live wsEndpoint into agent-browser --cdp ....
  7. Use scripts/discover_tabbit_cdp.py when you only need structured connection facts.
  8. Once connected, use the full normal agent-browser workflow for page operations.

Workflow

  1. For Tabbit requests, start by reading DevToolsActivePort directly or by running scripts/discover_tabbit_cdp.py.
  2. Return the connection facts the agent actually needs: activePortFile, port, browserPath, browserUrl, and wsEndpoint.
  3. Unless the user explicitly asks only for endpoint details, prefer scripts/run_agent_browser_on_tabbit.py immediately so the command becomes agent-browser --cdp <wsEndpoint> ....
  4. After that handoff, follow the normal agent-browser workflow for open, snapshot, click, fill, and other browser commands.
  5. If agent-browser is unavailable, say so plainly and surface the connection facts instead of inventing a custom CDP bridge.

Guidance

  • This skill solves the connection problem, not the general browser-operation problem.
  • Return structured connection data first, then any short explanatory note.
  • Prefer the lightest possible discovery path: DevToolsActivePort and the derived browser WebSocket endpoint.
  • Search the macOS Tabbit support directory first, then Tabbit Browser.
  • Prefer the full wsEndpoint over a raw port because Tabbit may not expose HTTP discovery routes.
  • Once a Tabbit task has started through run_agent_browser_on_tabbit.py, keep using that same wrapper path for the rest of the task unless the user explicitly asks otherwise.
  • Once connected, use standard agent-browser patterns for everything else.

Constraints

  • Do not assume a dedicated tabbit-devtools MCP server exists.
  • Do not assume the generic chrome-devtools session can be retargeted to Tabbit.
  • Do not turn this skill into a replacement for agent-browser.
  • Do not create a custom daemon, long-lived CDP proxy, or one-off WebSocket client for post-connection browser actions.
  • Do not promise that chrome-devtools MCP will automatically take over Tabbit.
  • If agent-browser cannot be launched in the current environment, stop at connection guidance and explain the limitation.
  • After connection, the browser workflow belongs to agent-browser, not to this skill.

Resources

相关 Skills

PPT处理

by anthropics

Universal
热门

处理 .pptx 全流程:创建演示文稿、提取和解析幻灯片内容、批量修改现有文件,支持模板套用、合并拆分、备注评论与版式调整。

涉及PPTX的创建、解析、修改到合并拆分都能一站搞定,连备注、模板和评论也能处理,做演示文稿特别省心。

效率与工作流
未扫描119.1k

技能工坊

by anthropics

Universal
热门

覆盖 Skill 从创建到迭代优化全流程:起草能力、补测试提示、跑评测与基准方差分析,并持续改写内容和描述,提升效果与触发准确率。

技能工坊把技能从创建、迭代到评测串成闭环,方差分析加描述优化,特别适合把触发准确率打磨得更稳。

效率与工作流
未扫描119.1k

Word文档

by anthropics

Universal
热门

覆盖Word/.docx文档的创建、读取、编辑与重排,适合生成报告、备忘录、信函和模板,也能处理目录、页眉页脚、页码、图片替换、查找替换、修订批注及内容提取整理。

搞定 .docx 的创建、改写与精排版,目录、批量替换、批注修订和图片更新都能自动化,做正式文档尤其省心。

效率与工作流
未扫描119.1k

相关 MCP 服务

文件系统

编辑精选

by Anthropic

热门

Filesystem 是 MCP 官方参考服务器,让 LLM 安全读写本地文件系统。

这个服务器解决了让 Claude 直接操作本地文件的痛点,比如自动整理文档或生成代码文件。适合需要自动化文件处理的开发者,但注意它只是参考实现,生产环境需自行加固安全。

效率与工作流
83.9k

by wonderwhy-er

热门

Desktop Commander 是让 AI 直接执行终端命令、管理文件和进程的 MCP 服务器。

这工具解决了 AI 无法直接操作本地环境的痛点,适合需要自动化脚本调试或文件批量处理的开发者。它能让你用自然语言指挥终端,但权限控制需谨慎,毕竟让 AI 执行 rm -rf 可不是闹着玩的。

效率与工作流
5.9k

EdgarTools

编辑精选

by dgunning

热门

EdgarTools 是无需 API 密钥即可解析 SEC EDGAR 财报的开源 Python 库。

这个工具解决了金融数据获取的痛点——直接让 AI 读取结构化财报,比如让 Claude 分析苹果的 10-K 文件。适合量化分析师或金融开发者快速构建数据管道。但注意,它依赖 SEC 网站稳定性,高峰期可能延迟。

效率与工作流
2.0k

评论