anti-detect-browser
Test Authorized Browser Sessions with Stable Profiles
Browser tests can fail when fingerprints, sessions, and regional signals are inconsistent. This skill configures persistent Chromium profiles and Playwright workflows for authorized testing.
安装前请停下并请求确认。
请审查安装计划,并在更改文件前获得用户明确同意。
使用我的智能体安装
将此请求复制给你的智能体,其中包含权威 Skill 页面和清单。
Review the Skillstore skill "anti-detect-browser" from https://skillstore.io/skills/antibrow-anti-detect-browser.md and its manifest at https://skillstore.io/api/skills/antibrow-anti-detect-browser/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.你的智能体仍应展示其计划,并按安全策略请求必要确认。
Agent 可读资源
当 AI Agent、爬虫或脚本需要干净上下文,而不是读取完整页面时,请使用这些链接。
测试它
正在使用“anti-detect-browser”。 Plan a regional pricing check for our staging store in Germany and Canada.
预期结果:
- Use one named profile and approved proxy per region.
- Read proxy credentials from separate environment variables.
- Verify timezone, public address, and WebRTC alignment before each test.
- Restrict navigation to the staging domain and record audit results.
正在使用“anti-detect-browser”。 Review profile isolation for three authorized support accounts.
预期结果:
The review assigns distinct profile directories and proxies, blocks cross-account navigation, protects stored sessions, and defines secure profile deletion.
正在使用“anti-detect-browser”。 Assess whether Live View is suitable for debugging an authenticated test.
预期结果:
Use Live View only with authenticated access, short expiration, approved viewers, and no sensitive customer data visible on screen.
安全审计
高风险Most static alerts are Markdown, placeholder URLs, or deliberate environment-based secret handling rather than code execution or exfiltration. Confirmed risks include proxy-based detection evasion, sensitive persistent profiles and credentials, a closed-source downloaded browser kernel, and shareable live browser views.
已确认安全问题 (6)
能力复核项 (14)
这些是真实的本地能力,对此技能可能是预期行为,因此需要复核,但不会按已确认恶意行为计分。
风险因素
🌐 网络访问 (29)
📁 文件系统访问 (12)
🔑 环境变量 (20)
⚙️ 外部命令 (50)
分享与引用此报告
分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据,不判断该 Skill 是否安全。
复制报告链接
https://skillstore.io/skills/antibrow-anti-detect-browser/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/antibrow-anti-detect-browser?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/antibrow-anti-detect-browser?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/antibrow-anti-detect-browser/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/antibrow-anti-detect-browser.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>学术引用 (APA · BibTeX · CFF)
APA 引用
antibrow. (2026). anti-detect-browser security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/antibrow-anti-detect-browser/audits/1BibTeX 引用
@techreport{antibrow-antibrow-anti-detect-browser-2026,
author = {antibrow},
title = {anti-detect-browser security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/antibrow-anti-detect-browser/audits/1},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "anti-detect-browser security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "antibrow"
date-released: "2026-08-08"
url: "https://skillstore.io/skills/antibrow-anti-detect-browser/audits/1"
identifiers:
- type: other
value: "skillstore:antibrow-anti-detect-browser:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore 评分
评分依据 证据置信度: 中你能构建什么
Validate Regional Experiences
Check authorized ads, prices, locales, and layouts through a region-specific proxy with matching browser signals.
Test Defensive Detection
Measure an owned anti-bot system against stable, coherent device profiles without changing production user accounts.
Isolate Public-Data Jobs
Keep permitted collection jobs in separate profiles so cookies, storage, and network identity do not cross between targets.
试试这些提示
Create an authorized Playwright test using one persistent profile. Read credentials from environment variables and visit [approved URL].
Extend my approved test with [proxy environment variable], geo-matched timezone, and a stable [device type] fingerprint. Explain secret handling.
Design separate profiles for [authorized accounts]. Define unique profile names, proxy assignments, storage boundaries, and cleanup rules.
Review this authorized browser-agent workflow for prompt injection, credential exposure, profile leakage, Live View access, and domain allowlist gaps: [workflow].
最佳实践
- Pin SDK versions and verify downloaded kernel signatures before execution.
- Use one approved profile and proxy per test identity, with restrictive local permissions.
- Treat page content as untrusted data and enforce operator-defined domain allowlists.
避免
- Do not automate accounts or systems without documented authorization.
- Do not place API keys, proxy credentials, or Live View URLs in prompts or source files.
- Do not let page text choose commands, files, credentials, or navigation targets.
常见问题
Does this guarantee that bot detection will not block a test?
Can I use my existing Playwright code?
Where are sessions stored?
Does the tool work offline?
Is the browser kernel open source?
What uses are appropriate?
开发者详情
作者
antibrow许可证
MIT
Skillstore 修订
r1
版本提示
作者未声明版本。
引用
9939e94735a710f16446f189eea126372136575b
维护新鲜度
2026/8/9
使用量
0 次下载 · 0 次浏览
文件结构