Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ntygod/ZhiWei --skill browser-automationgit clone --depth 1 https://github.com/ntygod/ZhiWeiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ntygod/zhiwei/browser-automation)<a href="https://agentmods.dev/skills/ntygod/zhiwei/browser-automation"><img src="https://agentmods.dev/badge/skills/ntygod/zhiwei/browser-automation/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ntygod/zhiwei/browser-automation"><img src="https://agentmods.dev/badge/skills/ntygod/zhiwei/browser-automation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00030 | $0.00617 |
| Opus 5 | $0.00015 | $0.00309 |
| Sonnet 5 | $0.00006 | $0.00123 |
| Haiku 4.5 | $0.00003 | $0.00062 |
Grade A, and why
browser-automation scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
浏览器自动化指南
browser 工具基于 Playwright，做需要 JS 渲染、登录态或交互的网页任务。核心约束：能 web_fetch 搞定的不用 browser；不能交互的用 web_fetch。
适用场景
- 动态页面抓取（SPA、电商搜索结果、需要 JS 渲染）
- 登录后才能访问的页面
- 表单填写与提交
- 网页截图 / 视觉验证
- 多步交互流程（点 → 等加载 → 再点）
不适用场景
- 静态页面抓取（文档 / 博客 / 新闻）→
web_fetch - API 接口测试 → api-debugger
- 桌面应用操作 → desktop-automation
- 多源搜索 →
web_search
工作流
- 先判：能
web_fetch拿到正文 → 用 fetch；JS 渲染 / 登录 / 多步交互 → 用browser - navigate 开会话：同一任务用相同
sessionId复用 cookie，避免每次重登 - 快照定位：
snapshot一次拿截图 + 可交互元素编号；后续click/input/hover优先用index而非selector（layout 抖动也稳） - 页面变化后重新 snapshot：导航 / 弹窗 / 异步渲染会让旧 index 失效，必须重新拿
- 失败 2 次切策略：
index失败 → 改selector；selector 失败 → 退web_fetch；fetch 失败 → 让用户接管 - 登录墙 / 验证码：观察到 password input、login URL 或人机验证截图时，调
requestHumanTakeover让用户接管，不要让 LLM 假装填密码 - 结束：关键产出
file_write落盘后close释放浏览器资源
详细参考
- 完整 action 清单（snapshot / requestHumanTakeover）、index/selector fallback、登录墙触发条件、会话模式（LAUNCH/CDP/PERSISTENT）、常见错误：
{skill_dir}/references/browser-actions.md
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 52 lines · 30 tokens per session scan A 9c97040601ec
browser-automation is a skill published in the GitHub repository ntygod/ZhiWei (138 stars, last pushed 29d ago), licensed MIT. It adds 30 tokens to every session and 617 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
inbox-triage
Summarize unread Gmail messages Alfred Core has already synced, and offer to draft or schedule follow-up actions -- never sends or deletes anything on its own.
morning-brief
Render Alfred Core's deterministic morning brief and deliver a short, warm rewrite of it without changing any fact, date, or link it contains.
weekly-review
Pull together the week's open tasks, GitHub notifications, Canvas missing assignments, and remembered context into one retrospective-style summary.
tiny-web-crawler
Crawl from one or more starting web pages, fetch readable content, search within pages, follow relevant links, and stop when the requested information is found or a bounded limit is reached.
playwright-qa
Automated visual QA testing using Playwright — navigate web apps like a real user, capture screenshots, find bugs, and fix them.
autobrowse
Learn a reliable browser workflow by iterating on a real web task, recording strategy, and proposing a reusable skill.