Token导航 LogoToken导航TokenDH.com
开发external-servicegithub未标认证来源可访问许可证需确认审计通过

security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

475

周安装

20

GitHub Stars

14

下载量

166
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:security(安全)
来源仓库:https://github.com/iankiku/forwward-teams
仓库路径:skills/security
安装命令:
npx skills add https://github.com/iankiku/forwward-teams --skill 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/iankiku/forwward-teams --skill security

简介

安全技能辅助安全审计、权限检查和漏洞排查。

  • 适合梳理敏感配置、检查依赖风险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。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。
  • 涉及密钥或生产系统时应先确认最小权限和操作边界。
  • security 属于开发类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Security — Compliance & Protection

Assume breach. Defense in depth. Least privilege everywhere.

Security Defaults

Every project ships with:

ControlImplementation
AuthOAuth 2.0 / OIDC via Auth.js or Supabase Auth
SessionsHTTP-only, Secure, SameSite=Strict cookies
Passwordsbcrypt/argon2, min 12 chars, no max limit
API authBearer tokens with expiry, refresh rotation
CORSExplicit allowlist, never * in production
HTTPSEverywhere. No exceptions. HSTS headers.
CSPContent-Security-Policy header on all pages
Rate limitingAuth endpoints: 5/min. API: 100/min. Adjust per use.

OWASP Top 10 Quick Reference

VulnerabilityPrevention
Injection (SQL, NoSQL, OS)Parameterized queries, ORMs, never string concat
Broken AuthMFA, session timeouts, account lockout
Sensitive Data ExposureEncrypt at rest + transit, minimize data collection
XXEDisable external entity processing
Broken Access ControlCheck permissions server-side on every request
Security MisconfigurationDefaults off, hardened configs, no debug in prod
XSSOutput encoding, CSP headers, sanitize HTML
Insecure DeserializationValidate and type-check all serialized data
Known Vulnerabilitiesnpm audit, pip audit, automated dependency updates
Insufficient LoggingLog auth events, access denied, input validation failures

HIPAA Compliance (Health-Tech)

Required if handling PHI (Protected Health Information):

RequirementImplementation
Encryption at restAES-256 for databases and file storage
Encryption in transitTLS 1.2+ everywhere
Access controlsRole-based, audit-logged, least privilege
Audit trailEvery PHI access logged with who, what, when
BAABusiness Associate Agreement with every vendor touching PHI
Data minimizationCollect only what's clinically necessary
Breach notification60-day notification requirement — have a plan
Employee trainingAnnual security awareness training

PHI includes: Names, dates, phone numbers, emails, SSN, medical record numbers, device IDs, biometric data, photos, and any data that could identify a patient.

Vendor checklist:

  • Does your cloud provider sign BAAs? (AWS, GCP, Azure: yes. Many others: no.)
  • Does your analytics tool see PHI? If yes, need BAA or strip PHI first.
  • Does your error tracking capture PHI in stack traces? Strip it.

SOC 2 Basics

Trust PrincipleWhat to Implement
SecurityAccess controls, encryption, firewalls, IDS
AvailabilityUptime monitoring, incident response, backups
Processing IntegrityInput validation, error handling, QA
ConfidentialityEncryption, access logging, data classification
PrivacyConsent, data retention, deletion, privacy policy

Start with: Security + Availability. Add others when customers require it.

Security Review Checklist

When reviewing code for security:

  1. Auth on every endpoint? — Not just the route, but the data query too
  2. Input validated at boundary? — Zod/Pydantic before any processing
  3. Secrets in env vars? — Never in code, git, or client bundles
  4. SQL parameterized? — No string concatenation, no template literals
  5. Error messages safe? — No stack traces, no internal paths to users
  6. Deps up to date?npm audit / pip audit clean
  7. Logging sufficient? — Auth events, permission failures, anomalies
  8. Data minimized? — Only collecting what's needed, only retaining as long as required

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

38.4%
按下载量换算64

Claude

28.94%
按下载量换算48

Cursor

19.32%
按下载量换算32

Gemini CLI

10.99%
按下载量换算18

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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。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