AutoBrowse
技能 已验证 活跃Self-improving browser automation via the auto-research loop. Iteratively runs a browsing task, reads the trace, and improves the navigation skill (strategy.md) until it reliably passes. Supports parallel runs across multiple tasks using sub-agents. Use when you want to build or improve browser automation skills for specific website tasks.
To build and refine reliable, production-ready browser automation skills for any website through an iterative, agent-driven process, significantly reducing costs and improving deterministic outcomes.
功能
- Iterative skill improvement via auto-research loop
- Parallel task execution using sub-agents
- Support for local and remote browser environments
- Generation of self-contained, graduated skills
- Clear documentation and examples for usage and setup
使用场景
- Building automated navigation for specific website tasks
- Improving existing brittle browser automation scripts
- Creating site-specific playbooks for agents
- Reducing LLM exploration costs for repetitive web tasks
非目标
- Executing arbitrary shell commands
- Replacing a full testing framework
- Managing the user's local browser installation
- Performing destructive operations outside the browser context
工作流
- Parse arguments and orient task mode (single/multi-task, iterations, env).
- Set up workspace (tasks/, traces/) and scaffold task.md/strategy.md if needed.
- Run the inner agent via `evaluate.mjs` for the specified task and environment.
- Read the trace (summary.md) to identify failures or areas for improvement.
- Formulate a single hypothesis and update strategy.md.
- Repeat the loop until the task passes consistently or max iterations are reached.
- If ready, graduate the task by writing a self-contained SKILL.md to ~/.claude/skills/.
- Generate a final report for multi-task sessions.
实践
- Iterative refinement
- Browser automation
- Agent development
先决条件
- Node.js 18+
- browse CLI
- ANTHROPIC_API_KEY
- BROWSERBASE_API_KEY (for remote env)
Trust
- info:Issues AttentionWith 2 issues opened and 0 closed in the last 90 days, maintainer responsiveness on issues is currently not demonstrable.
安装
请先添加 Marketplace
/plugin marketplace add browserbase/skills/plugin install skills@browserbase质量评分
已验证类似扩展
Agent Browser
100AI 代理的浏览器自动化 CLI。当用户需要与网站交互时使用，包括浏览页面、填写表单、点击按钮、截屏、提取数据、测试 Web 应用或自动化任何浏览器任务。触发条件包括请求“打开网站”、“填表”、“点击按钮”、“截屏”、“抓取页面数据”、“测试此 Web 应用”、“登录网站”、“自动化浏览器操作”或任何需要以编程方式进行 Web 交互的任务。
Chatgpt Search
100Search ChatGPT and extract the full response + hydration JSON that powers the UI. Attaches to a running Chrome instance (port 9222 by default), opens ChatGPT, submits a query, waits for the streamed response, and returns structured data: messages, product cards, hydration JSON, and API calls. Use when asked to "search chatgpt", "ask chatgpt", "chatgpt search", "get chatgpt response", or "scrape chatgpt".
Senior Backend Engineer
100Designs and implements backend systems including REST APIs, microservices, database architectures, authentication flows, and security hardening. Use when the user asks to "design REST APIs", "optimize database queries", "implement authentication", "build microservices", "review backend code", "set up GraphQL", "handle database migrations", or "load test APIs". Covers Node.js/Express/Fastify development, PostgreSQL optimization, API security, and backend architecture patterns.
Browser Extract
99Extract structured data via stored browser-templates or one-shot DOM queries, with mandatory AIDefence PII + prompt-injection gates before content reaches the model
Agent Browser
99通过浏览器自动化与 Slack 工作区进行交互。适用于用户需要检查未读频道、导航 Slack、发送消息、提取数据、查找信息、搜索对话或自动化任何 Slack 任务的场景。触发词包括“检查我的 Slack”、“有哪些未读频道”、“发送消息给”、“搜索 Slack”、“从 Slack 提取”、“查找谁说了”或任何需要程序化 Slack 交互的任务。
Dev Browser
99使用持久化页面状态进行浏览器自动化。当用户要求导航网站、填写表单、截屏、提取网页数据、测试 Web 应用或自动化浏览器工作流时使用。触发短语包括“转到 [url]”、“点击”、“填写表单”、“截屏”、“抓取”、“自动化”、“测试网站”、“登录”或任何浏览器交互请求。