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 Insajin/autopus-adk --skill browser-automationgit clone --depth 1 https://github.com/Insajin/autopus-adkWrote 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/insajin/autopus-adk/browser-automation)<a href="https://agentmods.dev/skills/insajin/autopus-adk/browser-automation"><img src="https://agentmods.dev/badge/skills/insajin/autopus-adk/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/insajin/autopus-adk/browser-automation"><img src="https://agentmods.dev/badge/skills/insajin/autopus-adk/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 | $0.00034 | $0.02717 |
| Opus 5 | $0.00017 | $0.01358 |
| Sonnet 5 | $0.00007 | $0.00543 |
| Haiku 4.5 | $0.00003 | $0.00272 |
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 11d 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 — 332 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Browser Automation Skill
터미널 환경을 자동 감지하여 cmux browser 또는 agent-browser CLI를 활용해 AI 에이전트가 직접 웹 페이지를 조작하고 검증하는 스킬입니다.
도구 선택
| 도구 | 용도 | 선택 기준 |
|---|---|---|
| cmux browser | cmux 환경 네이티브 조작 | cmux 터미널에서 자동 선택 (최우선) |
| agent-browser | 에이전트 직접 조작 | tmux/plain 환경 폴백 |
| Playwright | E2E 테스트 스위트 | 반복 실행, CI/CD, 회귀 테스트 |
백엔드는 auto terminal detect 결과에 따라 자동 선택됩니다. 수동 지정하지 마세요.
백엔드 자동 감지
브라우저 백엔드는 현재 터미널 환경에 따라 자동 선택됩니다.
감지 방법:
auto terminal detect
| 터미널 | 백엔드 | 명령어 접두사 |
|---|---|---|
| cmux | cmux browser | cmux browser <subcommand> |
| tmux / plain | agent-browser | agent-browser <subcommand> |
IMPORTANT: 스킬 실행 시 반드시 auto terminal detect를 먼저 호출하여 백엔드를 결정해야 합니다. agent-browser를 하드코딩하지 마세요.
명령어 매핑
| 동작 | cmux browser | agent-browser |
|---|---|---|
| 페이지 열기 | cmux browser open <url> |
agent-browser open <url> |
| 스냅샷 | cmux browser --surface <ref> snapshot |
agent-browser snapshot |
| 클릭 | cmux browser --surface <ref> click --selector <sel> |
agent-browser click <ref> |
| 입력 | cmux browser --surface <ref> fill --selector <sel> --text <text> |
agent-browser fill <ref> <text> |
| 스크린샷 | cmux browser --surface <ref> screenshot --out <path> |
agent-browser screenshot <path> |
| 대기 | cmux browser --surface <ref> wait --text <text> |
agent-browser wait --text <text> |
| URL 확인 | cmux browser --surface <ref> get url |
agent-browser get url |
| 텍스트 확인 | cmux browser --surface <ref> get text --selector <sel> |
agent-browser get text <ref> |
| 표시 여부 | cmux browser --surface <ref> is visible --selector <sel> |
agent-browser is visible <ref> |
cmux 고유 기능:
--surface <ref>: cmux browser는 surface 핸들로 세션을 관리합니다.open명령이 반환하는 surface ref를 후속 명령에 사용합니다.--snapshot-after: 클릭/입력 후 자동 스냅샷 (cmux browser --surface <ref> click --selector <sel> --snapshot-after)--interactive: 인터랙티브 스냅샷 모드 (cmux browser --surface <ref> snapshot --interactive)
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.
- 11d ago First seen · 332 lines · 34 tokens per session scan A 2dc06d8097ef
browser-automation is a skill published in the GitHub repository Insajin/autopus-adk (111 stars, last pushed today), licensed MIT. It adds 34 tokens to every session and 2,717 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
ui-testing
Create automated UI tests using Playwright for Streamlit and web applications. Use when writing end-to-end tests, automating UI testing, or testing new features.
playwright
Complete browser automation with Playwright. Auto-detects dev servers, writes clean test scripts to /tmp. Test pages, fill forms, take screenshots, check responsive design, validate UX, test login flows, check links, automate any browser task. Use when user wants to test websites, automate browser interactions…
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
hive.browser-automation
Required before any hive-browser CLI command. The browser is driven from the terminal by running hive-browser ... --json via terminalexec — not via MCP tools. Teaches the browser lifecycle rules (the bridge attaches to the USER'S running Chrome — never kill or launch browser processes; timeouts are transport issues…
hive.linkedin-automation
Read before automating LinkedIn with browser tools. LinkedIn combines shadow DOM (#interop-outlet), strict Trusted Types CSP that silently drops innerHTML, Lexical composer, native beforeunload dialogs that hang the bridge, and aggressive spam filters — each has bitten us at least once. Verified flows for profile…
hive.slack-notifications-setup
Set up a Slack notification channel (Sentinel) for a colony by driving the browser — reuse or create the "Hive Sentinel" Slack app from a JSON manifest, install it, capture the bot + app tokens, create/select the channel via the Slack API, and turn Sentinel on so the colony can ping the user on Slack and accept…