Token导航 LogoToken导航

Agent Skill · 研究检索

analyzing-dependencies(分析依赖关系)

analyzing-dependencies 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。

analyzing-dependencies 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,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 48,434。

总安装 689周安装 29GitHub Stars 2,127下载 241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
unknown

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:analyzing-dependencies(分析依赖关系)
来源仓库:https://github.com/jeremylongshore/claude-code-plugins-plus-skills
仓库路径:skills/analyzing-dependencies
安装命令:
npx skills add https://github.com/jeremylongshore/claude-code-plugins-plus-skills --skill analyzing-dependencies
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/jeremylongshore/claude-code-plugins-plus-skills --skill analyzing-dependencies

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。

用于扫描项目依赖中的安全漏洞和许可证合规问题。支持 npm、pip、Composer、Gem、Go 模块和 Cargo 生态。自动比对 CVE 数据库并提供升级修复建议。需访问项目目录及清单文件,至少一个包管理器 CLI 可用。依赖网络连接获取最新漏洞情报和元数据。

analyzing-dependencies 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/analyzing-dependencies

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/jeremylongshore/claude-code-plugins-plus-skill…;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

    它需要需要联网,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

analyzing-dependencies 怎么安装?

在终端执行 npx skills add https://github.com/jeremylongshore/claude-code-plugins-plus-skills --skill…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

analyzing-dependencies 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。

analyzing-dependencies 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

analyzing-dependencies 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

analyzing-dependencies 用的人多吗?

公开数据是累计安装 689 次,近一周安装 29 次,下载 241 次,GitHub 2,127 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。

analyzing-dependencies 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。
  • 来源仓库有 2,127 星,维护活跃度相对可查。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 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
analyzing-dependencies

分析依赖关系 · 本页

6892,127需要联网未明确有提醒

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

Analyzing Dependencies

Overview

Analyze project dependencies for known security vulnerabilities, outdated versions, and license compliance issues across multiple package ecosystems. This skill inspects npm, pip, Composer, Gem, Go module, and Cargo manifests and lock files, cross-references findings against CVE databases, and produces actionable remediation guidance with upgrade paths.

Prerequisites

  • Access to the target project directory and manifest files in ${CLAUDE_SKILL_DIR}/
  • At least one package manager CLI available: npm, pip/pip-audit, composer, gem, go, or cargo
  • Network access for querying vulnerability databases (NVD, GitHub Advisory Database, OSV)
  • Reference: ${CLAUDE_SKILL_DIR}/references/README.md for npm/pip audit report formats, license compatibility matrix, and dependency management best practices

Instructions

  1. Detect the project ecosystem by scanning ${CLAUDE_SKILL_DIR}/ for manifest files: package.json and package-lock.json (npm/Node.js), requirements.txt/pyproject.toml/Pipfile.lock (Python), composer.json/composer.lock (PHP), Gemfile/Gemfile.lock (Ruby), go.mod/go.sum (Go), Cargo.toml/Cargo.lock (Rust).
  2. For npm projects, run npm audit --json and parse the structured output. Map each advisory to its CVE identifier, CVSS score, severity level, vulnerable version range, and patched version.
  3. For Python projects, run pip-audit --format=json or parse safety check --json output. Cross-reference each vulnerability against the OSV database for additional context.
  4. For other ecosystems, run the equivalent audit command (composer audit, bundle audit, cargo audit, govulncheck) and normalize the output to a common finding format.
  5. Analyze the dependency tree for transitive vulnerabilities -- identify which direct dependency pulls in the vulnerable transitive dependency, and whether upgrading the direct dependency resolves the issue.
  6. Check for outdated packages by comparing installed versions against the latest available versions. Categorize updates as patch (safe), minor (likely safe), or major (breaking changes possible).
  7. Audit license compliance by extracting license declarations from each dependency. Flag packages using copyleft licenses (GPL, AGPL) in proprietary projects, packages with no declared license, and packages with license conflicts per the compatibility matrix in ${CLAUDE_SKILL_DIR}/references/README.md.
  8. Identify abandoned or unmaintained packages: flag dependencies with no releases in over 2 years, archived repositories, or known deprecation notices.
  9. Classify each finding by severity (critical, high, medium, low) using CVSS scores: critical >= 9.0, high >= 7.0, medium >= 4.0, low < 4.0.
  10. Generate a remediation plan with specific upgrade commands, alternative packages for abandoned dependencies, and a priority order based on severity and exploitability.

Output

  • Vulnerability report: Table with columns: Package, Installed Version, Vulnerability (CVE ID), CVSS Score, Severity, Patched Version, Direct/Transitive
  • Outdated packages: Table with columns: Package, Current Version, Latest Version, Update Type (patch/minor/major), Breaking Changes Risk
  • License audit: Table with columns: Package, License, Compatibility Status (OK, Warning, Conflict), Notes
  • Dependency tree visualization: For critical vulnerabilities, the chain from direct dependency to vulnerable transitive dependency
  • Remediation commands: Ready-to-run commands (e.g., npm install package@version, pip install --upgrade package==version) prioritized by severity
  • Executive summary: Total dependencies scanned, total vulnerabilities by severity, outdated count, license conflicts count

Error Handling

ErrorCauseSolution
npm audit returns exit code 1Vulnerabilities found (expected behavior)Parse the JSON output normally; exit code 1 indicates findings, not a tool failure
pip-audit not installedTool not available in the environmentInstall with pip install pip-audit or fall back to manual pip list --outdated combined with OSV API queries
Lock file missing or outdatedDependencies not properly lockedRun npm install, pip freeze, or equivalent to generate/update the lock file before scanning
Network timeout querying vulnerability DBFirewall or connectivity issueRetry with increased timeout; fall back to offline analysis of lock file versions against cached CVE data
Mixed ecosystem projectMultiple manifest files in one repoScan each ecosystem independently and combine results into a unified report
Private registry packages not foundAudit tools cannot resolve private packagesSkip private packages in the vulnerability scan; note them as "unverifiable" in the report

Examples

npm Pre-Deployment Audit

Run npm audit --json in ${CLAUDE_SKILL_DIR}/. Parse the output to identify critical and high severity advisories. For each, trace the dependency chain from direct dependency to vulnerable package. Produce upgrade commands: npm install express@4.19.2 to resolve CVE-2024-XXXXX in path-to-regexp. Flag any advisory without a fix available as requiring a workaround or alternative package.

Python Dependency Security Check

Run pip-audit --format=json -r ${CLAUDE_SKILL_DIR}/requirements.txt. Map each vulnerability to its CVE, CVSS score, and fixed version. For transitive dependencies, identify the direct dependency pulling in the vulnerable package. Recommend pinning to safe versions in requirements.txt and adding pip-audit to the CI pipeline.

License Compliance Scan

Extract licenses from ${CLAUDE_SKILL_DIR}/node_modules/ using license-checker --json or equivalent. Flag any GPL-3.0 or AGPL-3.0 licensed package used in a proprietary application as a license conflict. Flag packages with UNLICENSED or missing license fields as requiring legal review before production use.

Resources

平台分布

Codex

39.04%
按下载量换算94

Claude

28.84%
按下载量换算70

Cursor

19.3%
按下载量换算47

Gemini CLI

9.92%
按下载量换算24

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills