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 LinVireo/browsertap-mcp --skill browsertap-defaultgit clone --depth 1 https://github.com/LinVireo/browsertap-mcpWrote 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/linvireo/browsertap-mcp/browsertap-default)<a href="https://agentmods.dev/skills/linvireo/browsertap-mcp/browsertap-default"><img src="https://agentmods.dev/badge/skills/linvireo/browsertap-mcp/browsertap-default.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00053 | $0.09553 |
| Opus 5 | $0.00026 | $0.04777 |
| Sonnet 5 | $0.00011 | $0.01911 |
| Haiku 4.5 | $0.00005 | $0.00955 |
Grade A, and why
browsertap-default 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 254 lines — stays where its author put it; the contents beside it link to each section on GitHub.
浏览器任务只走 BTAP（browsertap-mcp）
browsertap-mcp (BTAP) 通过 CDP 桥连接真实已登录的 Chrome/Edge。使用本 skill 的浏览器任务应只使用 BTAP，不另起 headless、Playwright 或其他独立浏览器 profile。
为什么只保留 BTAP
- 复用真实登录态：当前 Chrome/Edge 的 cookie、SSO、会话都在。
- 默认部署面向单操作员；多 Agent 通过显式 session 与 owner 契约隔离。
- 桥故障时走 browsertap-bridge-recovery /
browsertap doctor，不换别的 browser 产品。
固定套路（照抄执行，禁止临场发挥）
模型翻车基本都发生在"自己解释错误语义"上。下面两类表是唯一正确响应，见到即做，不推理、不换路、不重试第二次。
高频流程模板（链路固定，步数固定）
| 场景 | 固定链路 |
|---|---|
| 只读抓取 | list_tabs → 借 U（记 original_url）或 open_new_tab → scan_page(session_id=...) → 需要数据再 execute_js(session_id=...) → 借用的还原 URL；owned 的带 owner_id 关 |
| 表单交互 | list_tabs → open_new_tab（记 session+generation+owner）→ wait_for_url → scan_page → page_type/page_click（每次带 session_id）→ wait_for_url → close_tabs(session_id, owner_id=...) |
| 下载附件 | download_file(url=..., session_id=...)。禁 ?dl=1、禁页面 fetch、禁裸 Page.navigate 猜目录 |
| 验证码页 | 同 tab page_click 一次 → challenge_stalled → 立刻交还用户，停手报告。不另起浏览器、不跑长 execute_js |
| 长页读取 | scan_page → scroll_page(to=...) → 再 scan_page（±5000px 窗口外内容必须滚动后再扫） |
错误语义 → 唯一动作
| 见到 | 唯一动作 |
|---|---|
no_response / switched_session |
list_tabs → switch_tab 重定位 → 只重试无副作用操作；有副作用脚本必须先 scan_page 确认未落地 |
busy |
等 10 秒重试一次；再 busy 就停手报告，不循环不删锁 |
blocked_by_dialog |
`handle_dialog(action="accept" |
challenge_stalled |
停，把 tab 交还用户 |
requires_user_action / input_activity_detected |
改用 page_*，或停手 |
cdp_timeout / debugger_detached |
list_tabs 验桥活 → 重试一次无副作用调用 |
debugger_conflict |
让用户关 DevTools/竞争 debugger，再原调用重试 |
401 / unauthorized |
转 [[browsertap-bridge-recovery]] 成因 6（token 文件不一致），不重启浏览器不重装扩展 |
Unknown command: downloads |
目标浏览器扩展旧了：chrome://extensions 手动 Reload，不重启桥 |
硬禁止（以前能跑、现在必炸）
execute_js里setTimeout/sleep 硬等 → 改wait_for/wait_for_url- 重放可能已执行的副作用脚本（
no_response后盲重试） - 对 Turnstile/登录盾页跑长
execute_js - 页面
fetch下载附件
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.
- 7d ago First seen · 254 lines · 53 tokens per session scan A 27f48a66c14b
browsertap-default is a skill published in the GitHub repository LinVireo/browsertap-mcp (2 stars, last pushed 5d ago), licensed MIT. It adds 53 tokens to every session and 9,553 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
owb
Open Web Bridge (OWB) — drive the user's own real browser with the owb command. Read pages behind their existing logins, gather and cross-check information, fill forms, walk multi-step flows, debug their site, audit responsive/accessibility behavior, and capture or reverse-engineer network traffic. Use this whenever…
browser-automation
Use this skill when the user asks to interact with a browser, take screenshots, inspect a page, capture network traffic, detect frameworks, click elements, fill forms, record browser sessions, or automate any browser task. Orchestrates Crawlio Browser's 150 browser tools via the search + execute + connecttab interface.
web-research
Structured web research, site audits, and competitive analysis. Acquire → Normalize → Analyze protocol with Evidence Mode findings.
clone
Capture a site's design system — colors, typography, spacing, layout, components — as structured findings.
compare
Side-by-side comparison of two websites across 11 dimensions. Produces Finding[] evidence per dimension.
dossier
Competitive intelligence dossier — compare a target against 2-4 competitors across all dimensions.