Token导航 LogoToken导航TokenDH.com
开发敏感数据github未标认证来源可访问许可证需确认审计提醒

astro-security天文安全

Agent Skill

用于辅助安全审计、权限检查、凭据风险、认证流程和常见漏洞排查。它适合让 Agent 梳理敏感配置、检查依赖风险、分析鉴权逻辑或生成安全复核清单。使用时不能把工具输出直接当最终结论,涉及密钥、令牌、用户数据或生产系统时,应先确认最小权限、脱敏方式和操作边界。

总安装

214

周安装

9

GitHub Stars

2

下载量

75
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:astro-security(天文安全)
来源仓库:https://github.com/soborbo/claudeskills
仓库路径:skills/astro-security
安装命令:
npx skills add https://github.com/soborbo/claudeskills --skill astro-security
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/soborbo/claudeskills --skill astro-security

简介

用于安全审计和常见漏洞排查。

  • 检查表单防护、服务器端验证和安全标头配置。
  • 禁止客户端暴露密钥,适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。
  • 必须实施服务端验证和速率限制。
  • astro-security 属于开发类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Astro Security Skill

Security patterns for lead generation sites.

Core Rules (Non-Negotiable)

ViolationResult
Production form without Turnstile + honeypotFAIL
Secret exposed client-sideFAIL
User input stored without server validationFAIL
Indexable staging environmentFAIL
Missing security headersFAIL
Cookie banner missing before analyticsFAIL

Form Security (Required)

Every form must have:

ProtectionImplementation
TurnstileCloudflare captcha (invisible mode)
HoneypotHidden field, reject if filled
Rate limitMax 5 submissions/IP/hour
ValidationServer-side Zod, never trust client
SanitizeStrip HTML, trim whitespace

See references/forms.md.

Security Headers (Required)

CSP Rules:

  • MUST disallow inline scripts unless hashed
  • MUST restrict script-src to required domains only
  • MUST test in report-only before enforcement

Add to _headers:

X-Content-Type-Options: nosniff
X-Frame-Options: DENY
X-XSS-Protection: 1; mode=block
Referrer-Policy: strict-origin-when-cross-origin
Permissions-Policy: camera=(), microphone=(), geolocation=()
Content-Security-Policy: [see references]
Strict-Transport-Security: max-age=31536000; includeSubDomains

See references/headers.md.

Environment Variables

# .env.example (commit this)
TURNSTILE_SITE_KEY=
TURNSTILE_SECRET_KEY=
RESEND_API_KEY=
GOOGLE_SHEETS_ID=

# .env (never commit)
# Add to .gitignore

Rules:

  • Never expose secrets client-side
  • Use import.meta.env for public vars only
  • Validate all env vars on build

Bot Protection

Cloudflare (free tier):

  • Bot Fight Mode: ON
  • Security Level: Medium
  • Challenge Passage: 30 minutes

Application level:

  • Turnstile on all forms
  • Honeypot fields
  • Rate limiting per IP
  • Block empty referrer (optional)

Third-Party Scripts

  • Use SRI (integrity hash) for CDN scripts
  • Load async/defer
  • Minimize scripts
  • Review GTM tags regularly

GDPR Compliance

Required:

  • Cookie banner (before non-essential cookies)
  • Privacy policy page
  • Form consent checkbox (if marketing)
  • Data retention policy
  • Right to deletion process

Cookie categories:

TypeConsentExamples
NecessaryNoSession, CSRF
AnalyticsYesGA4, Hotjar
MarketingYesMeta Pixel, Google Ads

See references/gdpr.md.

Input Validation

Never: Trust client-side alone, store raw input, render unsanitized HTML.

See forms.md for Zod schemas.

File Uploads

If needed: Max 5MB, whitelist types, rename files, store outside webroot.

Staging Protection

Password protect OR Cloudflare Access. Add noindex, block in robots.txt.

Error Handling

  • Error messages MUST NOT reveal stack traces or internals
  • API errors MUST return generic messages (Something went wrong)
  • Detailed errors allowed ONLY in development
  • 404/500 pages must not leak tech stack info

Dependencies

  • Minimize third-party scripts
  • Remove unused dependencies before launch
  • Review third-party access quarterly
  • Prefer self-hosted over CDN when possible

Definition of Done

Security requirements before launch:

  • Turnstile on all forms
  • Honeypot fields added
  • Rate limiting configured
  • Security headers set
  • HTTPS enforced
  • .env in.gitignore
  • No secrets in client code
  • Cookie banner working
  • Privacy policy linked
  • Staging protected
  • Error pages don't leak info

References

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.7%
按下载量换算29

Claude

28.96%
按下载量换算22

Cursor

18.84%
按下载量换算14

Gemini CLI

10.55%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills