Token导航 LogoToken导航

Agent Skill · 研究检索

owasp-checker(owasp 检查器)

owasp-checker 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

owasp-checker 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为接触敏感数据;许可证为 MIT,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 59,803。

总安装 376周安装 16GitHub Stars 21下载 132核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

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:owasp-checker(owasp 检查器)
来源仓库:https://github.com/matteocervelli/llms
仓库路径:skills/owasp-checker
安装命令:
npx skills add https://github.com/matteocervelli/llms --skill owasp-checker
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/matteocervelli/llms --skill owasp-checker

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

owasp-checker 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词、任务场景或来源线索快速定位候选结果。通过 npx skills add 命令从指定仓库安装,需结合原始 README 核验具体用法。

安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。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。

来源仓库内路径:skills/owasp-checker

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/matteocervelli/llms --skill owasp-checker;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

owasp-checker 怎么安装?

在终端执行 npx skills add https://github.com/matteocervelli/llms --skill owasp-checker。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

owasp-checker 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

owasp-checker 需要什么权限?

权限级别是接触敏感数据。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。

owasp-checker 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

owasp-checker 用的人多吗?

公开数据是累计安装 376 次,近一周安装 16 次,下载 132 次,GitHub 21 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

owasp-checker 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
owasp-checker

owasp 检查器 · 本页

37621接触敏感数据MIT通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

OWASP Top 10 Checker Skill

Purpose

This skill provides systematic verification of application compliance with the OWASP Top 10 2021 security standards, ensuring comprehensive security coverage across all critical categories.

When to Use

  • Final security validation before deployment
  • Security certification and compliance
  • Security audit preparation
  • Post-remediation verification
  • Quarterly security reviews
  • Pre-release security checklist

OWASP Top 10 2021 Compliance Workflow

A01:2021 - Broken Access Control

Risk: Users can act outside of their intended permissions

Compliance Checks:

1. Authorization Enforcement:

# Check for authorization decorators/middleware
grep -r "@requires_auth\|@login_required\|@permission_required" src/
grep -r "auth_required\|check_permission" src/

# Find routes without authorization
grep -r "@app.route\|@router.get\|@router.post" src/ --include="*.py" -A 5

Checklist:

  • Authorization enforced on every endpoint
  • Default deny access control
  • Server-side authorization (not client-side only)
  • Authorization checked on every request
  • Role-based access control (RBAC) implemented
  • Ownership verified for object access

2. Insecure Direct Object References (IDOR):

# Look for direct ID usage from requests
grep -r "request.*\['id'\]\|request.*\.id\|params\['id'\]" src/

Checklist:

  • No direct object references without validation
  • Ownership verified before access
  • Indirect references used (e.g., session-based)
  • UUID instead of sequential IDs where applicable

3. CORS Configuration:

# Check CORS settings
grep -r "Access-Control-Allow-Origin" src/ config/
grep -r "CORS.*origin" src/ --include="*.py" --include="*.js"

Checklist:

  • No wildcard (*) CORS unless absolutely necessary
  • Specific origin whitelist configured
  • Credentials mode properly configured
  • Preflight requests handled correctly

4. Disable Directory Listing:

# Check web server config
grep -r "autoindex\|directory.*listing" config/

Checklist:

  • Directory listing disabled
  • .git directory not accessible
  • Backup files not accessible

Status: ☐ Pass ☐ Fail


A02:2021 - Cryptographic Failures

Risk: Sensitive data exposed due to weak or missing encryption

Compliance Checks:

1. Data in Transit:

# Check TLS enforcement
grep -r "SECURE_SSL_REDIRECT\|HTTPS_ONLY\|ssl.*required" config/ src/
grep -r "tls.*version\|ssl.*version" config/

# Check for HTTP usage
grep -r "http://\|ws://" src/ | grep -v "localhost\|127.0.0.1"

Checklist:

  • TLS 1.2+ enforced
  • TLS 1.0/1.1 disabled
  • HTTPS redirect configured
  • HSTS header set (max-age >= 31536000)
  • Secure WebSocket (wss://) for real-time
  • Certificate validation enabled

2. Data at Rest:

# Check for encryption of sensitive data
grep -r "encrypt\|cipher\|AES" src/
grep -r "password.*plain\|password.*clear" src/

Checklist:

  • Passwords hashed (bcrypt/argon2/scrypt)
  • PII encrypted at rest
  • Database encryption enabled for sensitive columns
  • Backups encrypted
  • Key management system used

3. Weak Cryptography:

# Find weak algorithms
grep -r "md5\|sha1\|DES\|RC4" src/ --include="*.py" --include="*.js"
grep -r "ECB.*mode" src/

Checklist:

  • No MD5/SHA1 for security (only for checksums)
  • AES-256-GCM or ChaCha20-Poly1305 for encryption
  • Argon2id or bcrypt for passwords
  • SHA-256/SHA-3 for hashing
  • Proper IV/nonce generation
  • No ECB mode

4. Random Number Generation:

# Check RNG usage
grep -r "random\.random\|Math\.random" src/
grep -r "secrets\|os\.urandom\|crypto\.randomBytes" src/

Checklist:

  • Cryptographically secure RNG (secrets, os.urandom)
  • No Math.random() or random.random() for security
  • Sufficient entropy for keys/tokens

Status: ☐ Pass ☐ Fail


A03:2021 - Injection

Risk: Untrusted data sent to interpreter as command/query

Compliance Checks:

1. SQL Injection Prevention:

# Find string concatenation in SQL
grep -r "execute.*%\|execute.*\+\|execute.*format\|execute.*f\"" src/ --include="*.py"
grep -r "SELECT.*\+\|INSERT.*\+\|UPDATE.*\+\|DELETE.*\+" src/

Checklist:

  • Parameterized queries/prepared statements
  • ORM used correctly (no raw SQL with user input)
  • No string concatenation in SQL
  • Input validation on all parameters
  • Least privilege database accounts

2. Command Injection Prevention:

# Find shell command execution
grep -r "subprocess.*shell=True\|os\.system\|os\.popen" src/ --include="*.py"
grep -r "exec\|eval\|child_process" src/ --include="*.js"

Checklist:

  • No shell=True with user input
  • Command arguments as list, not string
  • Input validation/sanitization
  • Whitelist allowed commands
  • No eval() or exec() with user input

3. LDAP Injection:

grep -r "ldap.*search\|ldap.*filter" src/

Checklist:

  • LDAP queries parameterized
  • Special characters escaped
  • Input validation

4. NoSQL Injection:

grep -r "find.*\$where\|\.exec(" src/ --include="*.js"

Checklist:

  • MongoDB operators sanitized
  • $where clauses avoided or validated
  • Input type validation

5. Template Injection:

grep -r "render_template_string\|Jinja2.*from_string" src/
grep -r "autoescape.*False" src/

Checklist:

  • No user input in template compilation
  • Auto-escaping enabled
  • Safe template rendering

Status: ☐ Pass ☐ Fail


A04:2021 - Insecure Design

Risk: Missing or ineffective security controls in design

Compliance Checks:

1. Threat Modeling:

Checklist:

  • Threat model documented
  • Attack surface analyzed
  • Trust boundaries identified
  • Data flow diagrams created
  • Security requirements defined

2. Security Design Patterns:

Checklist:

  • Defense in depth implemented
  • Fail securely (errors don't expose data)
  • Least privilege principle
  • Separation of duties
  • Complete mediation (check every access)

3. Rate Limiting:

grep -r "rate.*limit\|throttle" src/ config/

Checklist:

  • Rate limiting on authentication endpoints
  • Rate limiting on API endpoints
  • Account lockout after failed attempts
  • CAPTCHA for public forms

4. Business Logic:

Checklist:

  • Transaction integrity enforced
  • Workflow state validated
  • Resource limits defined
  • Input bounds checked
  • Edge cases handled

Status: ☐ Pass ☐ Fail


A05:2021 - Security Misconfiguration

Risk: Insecure default configurations, incomplete setups

Compliance Checks:

1. Security Headers:

# Check for security headers
grep -r "X-Frame-Options\|X-Content-Type-Options\|Content-Security-Policy" src/ config/
grep -r "Strict-Transport-Security\|X-XSS-Protection" src/ config/

Checklist:

  • X-Frame-Options: DENY or SAMEORIGIN
  • X-Content-Type-Options: nosniff
  • Content-Security-Policy configured
  • Strict-Transport-Security: max-age=31536000
  • X-XSS-Protection: 1; mode=block
  • Referrer-Policy: strict-origin-when-cross-origin

2. Error Handling:

# Check debug mode
grep -r "DEBUG.*True\|development.*mode" config/ src/
grep -r "traceback\|stack.*trace" src/

Checklist:

  • Debug mode disabled in production
  • Generic error messages to users
  • Detailed errors logged, not displayed
  • No stack traces exposed
  • Custom error pages configured

3. Default Credentials:

# Find hardcoded credentials
grep -r "password.*=.*admin\|password.*=.*password" src/ config/

Checklist:

  • All default credentials changed
  • No hardcoded passwords
  • Credentials in environment variables
  • Secrets management system used

4. Unnecessary Features:

# Check for sample/test code
find . -name "*sample*" -o -name "*test*" -o -name "*demo*" | grep -v node_modules

Checklist:

  • Sample code removed
  • Unused endpoints disabled
  • Test accounts removed
  • Development tools disabled in production
  • Unnecessary services stopped

5. Missing Patches:

# Check dependency status
pip list --outdated 2>/dev/null || npm outdated 2>/dev/null

Checklist:

  • Dependencies up to date
  • Security patches applied
  • Regular update schedule
  • Vulnerability monitoring enabled

Status: ☐ Pass ☐ Fail


A06:2021 - Vulnerable and Outdated Components

Risk: Using components with known vulnerabilities

Compliance Checks:

1. Dependency Inventory:

# List all dependencies
pip list --format=json > dependencies.json 2>/dev/null || npm list --json > dependencies.json 2>/dev/null

Checklist:

  • Complete dependency inventory (SBOM)
  • Direct and transitive dependencies tracked
  • License compliance verified
  • Dependency sources trusted

2. Vulnerability Scanning:

# Scan for vulnerabilities
pip-audit --format json 2>/dev/null || npm audit --json 2>/dev/null
safety check --json 2>/dev/null

Checklist:

  • Regular vulnerability scans (weekly minimum)
  • No critical vulnerabilities
  • High vulnerabilities remediated
  • Vulnerability remediation SLA defined

3. Version Management:

# Check for version pinning
cat requirements.txt setup.py package.json 2>/dev/null

Checklist:

  • Versions pinned in lockfiles
  • Compatible version ranges defined
  • Regular updates scheduled
  • Breaking changes reviewed before update

4. Component Retirement:

Checklist:

  • No unmaintained dependencies
  • EOL software replaced
  • Deprecated features not used
  • Migration plan for aging components

Status: ☐ Pass ☐ Fail


A07:2021 - Identification and Authentication Failures

Risk: Weak authentication allows impersonation

Compliance Checks:

1. Password Policy:

# Check password requirements
grep -r "password.*length\|password.*complexity" src/
grep -r "MIN_PASSWORD_LENGTH\|PASSWORD_VALIDATORS" config/ src/

Checklist:

  • Minimum 8 characters (12+ recommended)
  • Complexity requirements enforced
  • Common passwords blocked
  • Password strength meter implemented
  • Password history (no reuse)

2. Multi-Factor Authentication:

grep -r "mfa\|2fa\|totp\|two.*factor" src/

Checklist:

  • MFA available for sensitive operations
  • MFA enforced for admin accounts
  • TOTP/hardware token support
  • Backup codes provided

3. Session Management:

# Check session configuration
grep -r "SESSION.*TIMEOUT\|session.*expir" config/ src/
grep -r "SESSION_COOKIE_SECURE\|SESSION_COOKIE_HTTPONLY" config/ src/

Checklist:

  • Session timeout configured (15-30 min idle)
  • Absolute session timeout (e.g., 8 hours)
  • Session invalidated on logout
  • Secure session cookies (Secure, HttpOnly, SameSite)
  • Session fixation prevented (regenerate on login)
  • Concurrent session limits

4. Credential Storage:

# Check password hashing
grep -r "bcrypt\|argon2\|scrypt\|pbkdf2" src/
grep -r "hashlib\.md5.*password\|hashlib\.sha1.*password" src/

Checklist:

  • Passwords hashed with bcrypt/argon2/scrypt
  • Salt unique per password
  • Work factor appropriate (bcrypt: 12+, argon2: per OWASP)
  • No reversible encryption for passwords

5. Account Enumeration:

# Check error messages
grep -r "user.*not.*found\|invalid.*username" src/

Checklist:

  • Generic authentication errors ("Invalid credentials")
  • Same response time for valid/invalid users
  • Password reset doesn't confirm email existence
  • Registration doesn't confirm email existence

6. Brute Force Protection:

grep -r "login.*attempt\|failed.*attempt\|account.*lock" src/

Checklist:

  • Account lockout after N failed attempts (5-10)
  • Temporary lockout (not permanent)
  • Rate limiting on login endpoint
  • CAPTCHA after failed attempts
  • Login attempt logging

Status: ☐ Pass ☐ Fail


A08:2021 - Software and Data Integrity Failures

Risk: Code and infrastructure not protected from integrity violations

Compliance Checks:

1. Insecure Deserialization:

# Check for unsafe deserialization
grep -r "pickle\.loads\|yaml\.load\(" src/ --include="*.py"
grep -r "eval\|unserialize" src/

Checklist:

  • No pickle with untrusted data
  • yaml.safe_load() used (not yaml.load())
  • JSON preferred over pickle
  • Deserialization input validated
  • No eval() with untrusted data

2. CI/CD Pipeline Security:

Checklist:

  • Pipeline configuration in version control
  • Code signing for releases
  • Secret scanning in CI/CD
  • Dependency verification
  • Build reproducibility
  • Artifact signing
  • Deployment approval process

3. Update Mechanism:

Checklist:

  • Updates delivered over HTTPS
  • Update signatures verified
  • Automatic updates signed
  • Rollback mechanism available
  • Update integrity checked

4. Supply Chain Security:

Checklist:

  • Dependencies from trusted sources
  • Dependency hash verification
  • Software Bill of Materials (SBOM)
  • Third-party code reviewed
  • Vendor security assessment

Status: ☐ Pass ☐ Fail


A09:2021 - Security Logging and Monitoring Failures

Risk: Breaches not detected, incidents not responded to

Compliance Checks:

1. Security Event Logging:

# Check logging implementation
grep -r "logging\|logger\|log\." src/ --include="*.py" --include="*.js"
grep -r "audit.*log\|security.*log" src/

Checklist:

  • Authentication events logged (success/failure)
  • Authorization failures logged
  • Input validation failures logged
  • Sensitive operations logged
  • Administrative actions logged
  • Access to sensitive data logged

2. Log Content:

Checklist:

  • Timestamp (UTC)
  • User/session identifier
  • Action performed
  • Resource accessed
  • Source IP address
  • Outcome (success/failure)
  • No sensitive data in logs (passwords, tokens, PII)

3. Log Protection:

# Check log file permissions
find . -name "*.log" -ls 2>/dev/null

Checklist:

  • Log files write-only for application
  • Logs protected from tampering
  • Logs rotated regularly
  • Log retention policy defined
  • Logs backed up
  • Centralized logging implemented

4. Monitoring and Alerting:

Checklist:

  • Real-time security monitoring
  • Automated alerting configured
  • Failed login threshold alerts
  • Unusual activity detection
  • Anomaly detection
  • Security dashboard available

5. Incident Response:

Checklist:

  • Incident response plan documented
  • Response team identified
  • Alert escalation procedures
  • Incident logging and tracking
  • Post-incident review process

Status: ☐ Pass ☐ Fail


A10:2021 - Server-Side Request Forgery (SSRF)

Risk: Application fetches remote resources without validating user-supplied URL

Compliance Checks:

1. URL Validation:

# Find URL fetching code
grep -r "requests\.get\|urllib\.request\|fetch\|axios\.get" src/
grep -r "url.*request\|user.*url" src/

Checklist:

  • User-supplied URLs validated
  • URL whitelist implemented
  • Protocol whitelist (HTTP/HTTPS only)
  • Private IP ranges blocked
  • DNS rebinding protection
  • URL redirection limited

2. Network Segmentation:

Checklist:

  • Application in DMZ
  • Internal services not accessible from app
  • Firewall rules deny by default
  • Outbound traffic restricted
  • Service mesh/network policies configured

3. Input Validation:

# Check for URL sanitization
grep -r "validate.*url\|sanitize.*url\|parse.*url" src/

Checklist:

  • URL parsing and validation
  • Hostname validation
  • Port restrictions
  • Path normalization
  • No file:// protocol
  • No access to metadata services (169.254.169.254)

Status: ☐ Pass ☐ Fail


Overall Compliance Report Format

# OWASP Top 10 2021 Compliance Report

**Date**: [YYYY-MM-DD]
**Application**: [name]
**Assessed By**: OWASP Checker

## Compliance Summary

| Category | Status | Critical Issues | Notes |
|----------|--------|-----------------|-------|
| A01 - Broken Access Control | ✅/⚠️/❌ | [count] | [summary] |
| A02 - Cryptographic Failures | ✅/⚠️/❌ | [count] | [summary] |
| A03 - Injection | ✅/⚠️/❌ | [count] | [summary] |
| A04 - Insecure Design | ✅/⚠️/❌ | [count] | [summary] |
| A05 - Security Misconfiguration | ✅/⚠️/❌ | [count] | [summary] |
| A06 - Vulnerable Components | ✅/⚠️/❌ | [count] | [summary] |
| A07 - Auth Failures | ✅/⚠️/❌ | [count] | [summary] |
| A08 - Integrity Failures | ✅/⚠️/❌ | [count] | [summary] |
| A09 - Logging Failures | ✅/⚠️/❌ | [count] | [summary] |
| A10 - SSRF | ✅/⚠️/❌ | [count] | [summary] |

**Legend**:
- ✅ Pass: Fully compliant
- ⚠️ Partial: Some issues, not critical
- ❌ Fail: Critical issues found

**Overall Compliance**: [XX]% ([X]/10 categories passed)

## Critical Findings

[List all critical non-compliance items that must be fixed]

## Recommendations

### Immediate (Critical)
1. [Item]

### Short-term (High)
1. [Item]

### Long-term (Medium)
1. [Item]

## Certification

This application [IS / IS NOT] compliant with OWASP Top 10 2021 standards.

**Assessor**: [name]
**Date**: [YYYY-MM-DD]
**Next Assessment**: [YYYY-MM-DD]

Best Practices

Assessment Process:

  • Complete all categories
  • Document evidence for each check
  • Retest after remediation
  • Keep assessment current (quarterly)

Compliance Tracking:

  • Track compliance over time
  • Maintain compliance dashboard
  • Regular reassessment (quarterly)
  • Update after major changes

Remediation:

  • Fix critical items immediately
  • Schedule high/medium items
  • Document compensating controls
  • Verify fixes effectiveness

Documentation:

  • Keep detailed compliance records
  • Document exceptions with justification
  • Track remediation progress
  • Maintain audit trail

Integration with Security Workflow

Input: Implemented and tested application Process: Systematic OWASP Top 10 compliance verification Output: Compliance report with certification status Next Step: Security certification or deployment approval


Remember

  • Compliance is not security: OWASP Top 10 is baseline, not complete security
  • Context matters: Adapt checks to your application type
  • Defense in depth: Multiple layers of controls
  • Continuous compliance: Not a one-time check
  • Document everything: Maintain audit trail
  • Stay current: OWASP Top 10 updates periodically

Your goal is to ensure applications meet OWASP Top 10 security standards and maintain compliance over time.

平台分布

OpenCode

25.96%
按下载量换算34

Antigravity

21.56%
按下载量换算28

Claude Code

18.06%
按下载量换算24

Codex

13.27%
按下载量换算18

Gemini CLI

7.73%
按下载量换算10

windsurf

3.18%
按下载量换算4

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

接触敏感数据

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。

继续浏览同类 Skills