Token导航 LogoToken导航TokenDH.com
研究检索external-servicegithub未标认证来源可访问clear审计通过

managing-vulnerabilities管理漏洞

Agent Skill

managing-vulnerabilities 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。

总安装

541

周安装

23

GitHub Stars

350

下载量

190
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

3

许可证

MIT

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:managing-vulnerabilities(管理漏洞)
来源仓库:https://github.com/ancoleman/ai-design-components
仓库路径:skills/managing-vulnerabilities
安装命令:
npx skills add https://github.com/ancoleman/ai-design-components --skill managing-vulnerabilities
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/ancoleman/ai-design-components --skill managing-vulnerabilities

简介

用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。

  • 适用于需要根据关键词或任务场景进行信息检索的场景,如漏洞管理。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装,需确认权限范围和是否触发联网操作。
  • 建议结合原始 README 核验具体用法,注意检查维护状态和潜在的文件读写行为。
  • 安装前应评估是否需要执行外部命令或访问敏感数据,避免误改生产环境。

SKILL.md

Vulnerability Management

Implement comprehensive vulnerability detection and remediation workflows across containers, source code, dependencies, and running applications. This skill covers multi-layer scanning strategies, SBOM generation (CycloneDX and SPDX), risk-based prioritization using CVSS/EPSS/KEV, and CI/CD security gate patterns.

When to Use This Skill

Invoke this skill when:

  • Building security scanning into CI/CD pipelines
  • Generating Software Bills of Materials (SBOMs) for compliance
  • Prioritizing vulnerability remediation using risk-based approaches
  • Implementing security gates (fail builds on critical vulnerabilities)
  • Scanning container images before deployment
  • Detecting secrets, misconfigurations, or code vulnerabilities
  • Establishing DevSecOps practices and automation
  • Meeting regulatory requirements (SBOM mandates, Executive Order 14028)

Multi-Layer Scanning Strategy

Vulnerability management requires scanning at multiple layers. Each layer detects different types of security issues.

Layer Overview

Container Image Scanning

  • Detects vulnerabilities in OS packages, language dependencies, and binaries
  • Tools: Trivy (comprehensive), Grype (accuracy-focused), Snyk Container (commercial)
  • When: Every container build, base image selection, registry admission control

SAST (Static Application Security Testing)

  • Analyzes source code for security flaws before runtime
  • Tools: Semgrep (fast, semantic), Snyk Code (developer-first), SonarQube (enterprise)
  • When: Every commit, PR checks, main branch protection

DAST (Dynamic Application Security Testing)

  • Tests running applications for vulnerabilities (black-box testing)
  • Tools: OWASP ZAP (open-source), StackHawk (CI/CD native), Burp Suite (manual + automated)
  • When: Staging environment testing, API validation, authentication testing

SCA (Software Composition Analysis)

  • Analyzes third-party dependencies for known vulnerabilities
  • Tools: Dependabot (GitHub native), Renovate (advanced), Snyk Open Source (commercial)
  • When: Every build, dependency updates, license audits

Secret Scanning

  • Prevents secrets from being committed to source code
  • Tools: Gitleaks (fast, configurable), TruffleHog (entropy detection), GitGuardian (commercial)
  • When: Pre-commit hooks, repository scanning, CI/CD artifact checks

Quick Tool Selection

Container Image → Trivy (default choice) OR Grype (accuracy focus)
Source Code → Semgrep (open-source) OR Snyk Code (commercial)
Running Application → OWASP ZAP (open-source) OR StackHawk (CI/CD native)
Dependencies → Dependabot (GitHub) OR Renovate (advanced automation)
Secrets → Gitleaks (open-source) OR GitGuardian (commercial)

For detailed tool selection guidance, see references/tool-selection.md.

SBOM Generation

Software Bills of Materials (SBOMs) provide a complete inventory of software components and dependencies. Required for compliance and security transparency.

CycloneDX vs. SPDX

CycloneDX (Recommended for DevSecOps)

  • Security-focused, OWASP-maintained
  • Native vulnerability references
  • Fast, lightweight (JSON/XML/ProtoBuf)
  • Best for: DevSecOps pipelines, vulnerability tracking

SPDX (Recommended for Legal/Compliance)

  • License compliance focus, ISO standard (ISO/IEC 5962:2021)
  • Comprehensive legal metadata
  • Government/defense preferred format
  • Best for: Legal teams, compliance audits, federal requirements

Generating SBOMs

With Trivy (CycloneDX or SPDX):

# CycloneDX format (recommended for security)
trivy image --format cyclonedx --output sbom.json myapp:latest

# SPDX format (for compliance)
trivy image --format spdx-json --output sbom-spdx.json myapp:latest

# Scan SBOM (faster than re-scanning image)
trivy sbom sbom.json --severity HIGH,CRITICAL

With Syft (high accuracy):

# Generate CycloneDX
syft myapp:latest -o cyclonedx-json=sbom.json

# Generate SPDX
syft myapp:latest -o spdx-json=sbom-spdx.json

# Pipe to Grype for scanning
syft myapp:latest -o json | grype

For comprehensive SBOM patterns and storage strategies, see references/sbom-guide.md.

Vulnerability Prioritization

Not all vulnerabilities require immediate action. Prioritize based on actual risk using CVSS, EPSS, and KEV.

Modern Risk-Based Prioritization

Step 1: Gather Metrics

MetricSourcePurpose
CVSS Base ScoreNVD, vendor advisoriesVulnerability severity (0-10)
EPSS ScoreFIRST.org APIExploitation probability (0-1)
KEV StatusCISA KEV CatalogActively exploited CVEs
Asset CriticalityInternal CMDBBusiness impact if compromised
ExposureNetwork topologyInternet-facing vs. internal

Step 2: Calculate Priority

Priority Score = (CVSS × 0.3) + (EPSS × 100 × 0.3) + (KEV × 50) + (Asset × 0.2) + (Exposure × 0.2)

KEV: 1 if in KEV catalog, 0 otherwise
Asset: 1 (Critical), 0.7 (High), 0.4 (Medium), 0.1 (Low)
Exposure: 1 (Internet-facing), 0.5 (Internal), 0.1 (Isolated)

Step 3: Apply SLA Tiers

PriorityCriteriaSLAAction
P0 - CriticalKEV + Internet-facing + Critical asset24 hoursEmergency patch immediately
P1 - HighCVSS ≥ 9.0 OR (CVSS ≥ 7.0 AND EPSS ≥ 0.1)7 daysPrioritize in sprint, patch ASAP
P2 - MediumCVSS 7.0-8.9 OR EPSS ≥ 0.0530 daysNormal sprint planning
P3 - LowCVSS 4.0-6.9, EPSS < 0.0590 daysBacklog, maintenance windows
P4 - InfoCVSS < 4.0No SLATrack, address opportunistically

Example: Log4Shell (CVE-2021-44228)

CVSS: 10.0
EPSS: 0.975 (97.5% exploitation probability)
KEV: Yes (CISA catalog)
Asset: Critical (payment API)
Exposure: Internet-facing

Priority Score = (10 × 0.3) + (97.5 × 0.3) + 50 + (1 × 0.2) + (1 × 0.2) = 82.65
Result: P0 - Critical (24-hour SLA)

For complete prioritization framework and automation scripts, see references/prioritization-framework.md.

CI/CD Integration Patterns

Multi-Stage Security Pipeline

Implement progressive security gates across pipeline stages:

Stage 1: Pre-Commit (Developer Workstation)

Tools: Secret scanning (Gitleaks), SAST (Semgrep)
Threshold: Block high-confidence secrets, critical SAST findings
Speed: < 10 seconds

Stage 2: Pull Request (CI Pipeline)

Tools: SAST, SCA, Secret scanning
Threshold: No Critical/High vulnerabilities, no secrets
Speed: < 5 minutes
Action: Block PR merge until fixed

Stage 3: Build (CI Pipeline)

Tools: Container scanning (Trivy), SBOM generation
Threshold: No Critical vulnerabilities in production dependencies
Artifacts: SBOM stored, scan results uploaded
Speed: < 2 minutes
Action: Fail build on Critical findings

Stage 4: Pre-Deployment (Staging)

Tools: DAST, Integration tests
Threshold: No Critical/High DAST findings
Speed: 10-30 minutes
Action: Gate deployment to production

Stage 5: Production (Runtime)

Tools: Continuous scanning, runtime monitoring
Threshold: Alert on new CVEs in deployed images
Action: Alert security team, plan patching

Example: GitHub Actions Multi-Stage Scan

name: Security Scan Pipeline

on: [push, pull_request]

jobs:
  secrets:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: trufflesecurity/trufflehog@main
        with:
          path: ./
          extra_args: --only-verified

  sast:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: semgrep/semgrep-action@v1
        with:
          config: p/security-audit

  container:
    runs-on: ubuntu-latest
    needs: [secrets, sast]
    steps:
      - uses: actions/checkout@v4
      - run: docker build -t myapp:${{ github.sha }} .

      - uses: aquasecurity/trivy-action@master
        with:
          image-ref: myapp:${{ github.sha }}
          format: sarif
          output: trivy-results.sarif
          severity: HIGH,CRITICAL
          exit-code: 1

      - name: Generate SBOM
        run: |
          trivy image --format cyclonedx \
            --output sbom.json myapp:${{ github.sha }}

      - uses: actions/upload-artifact@v3
        with:
          name: sbom
          path: sbom.json

For complete CI/CD patterns (GitLab CI, Jenkins, Azure Pipelines), see references/ci-cd-patterns.md.

Container Scanning with Trivy

Trivy is the recommended default for container scanning: comprehensive, fast, and CI/CD native.

Basic Usage

# Scan container image
trivy image alpine:latest

# Scan with severity filter
trivy image --severity HIGH,CRITICAL alpine:latest

# Fail on findings (CI/CD)
trivy image --exit-code 1 --severity HIGH,CRITICAL myapp:latest

# Generate SBOM
trivy image --format cyclonedx --output sbom.json alpine:latest

# Scan filesystem
trivy fs /path/to/project

# Scan Kubernetes manifests
trivy config deployment.yaml

Configuration (.trivy.yaml)

severity: HIGH,CRITICAL
exit-code: 1
ignore-unfixed: true  # Only fail on fixable vulnerabilities
vuln-type: os,library
skip-dirs:
  - node_modules
  - vendor
ignorefile: .trivyignore

Ignoring False Positives (.trivyignore)

# False positive
CVE-2023-12345

# Accepted risk with justification
CVE-2023-67890  # Risk accepted: Not exploitable in our use case

# Development dependency (not in production)
CVE-2023-11111  # Dev dependency only

GitHub Actions Integration

- name: Trivy Scan
  uses: aquasecurity/trivy-action@master
  with:
    image-ref: myapp:${{ github.sha }}
    format: sarif
    output: trivy-results.sarif
    severity: HIGH,CRITICAL
    exit-code: 1

- name: Upload to GitHub Security
  uses: github/codeql-action/upload-sarif@v2
  if: always()
  with:
    sarif_file: trivy-results.sarif

Alternative: Grype for Accuracy

Grype focuses on minimal false positives and works with Syft for SBOM generation.

Important: Use Grype v0.104.1 or later (credential disclosure CVE-2025-65965 patched in earlier versions).

Basic Usage

# Scan container image
grype alpine:latest

# Scan with severity threshold
grype alpine:latest --fail-on high

# Scan SBOM (faster)
grype sbom:./sbom.json

# Syft + Grype workflow
syft alpine:latest -o json | grype --fail-on critical

When to Use Grype

  • Projects sensitive to false positives
  • SBOM-first workflows (generate with Syft, scan with Grype)
  • Need second opinion validation
  • Anchore ecosystem users

For complete tool comparisons and selection criteria, see references/tool-selection.md.

Security Gates and Thresholds

Progressive Threshold Strategy

Balance security and development velocity with progressive gates. Configure different thresholds for PR checks (fast, HIGH+CRITICAL), builds (comprehensive), and deployments (strict, CRITICAL only).

Policy-as-Code

Use OPA (Open Policy Agent) for automated policy enforcement. Create policies to deny Critical vulnerabilities, enforce KEV catalog checks, and implement environment-specific rules.

For complete policy patterns, baseline detection, and OPA examples, see references/policy-as-code.md.

Remediation Workflows

Automated Remediation

Set up automated workflows to scan daily, extract fixable vulnerabilities, update dependencies, and create remediation pull requests automatically.

SLA Tracking

Track vulnerability remediation against SLA targets (P0: 24 hours, P1: 7 days, P2: 30 days, P3: 90 days). Monitor overdue vulnerabilities and escalate as needed.

False Positive Management

Maintain suppression files (.trivyignore) with documented justifications, review dates, and approval tracking. Implement workflows for false positive triage and approval.

For complete remediation workflows, SLA trackers, and automation scripts, see references/remediation-workflows.md.

Integration with Related Skills

building-ci-pipelines

  • Add security stages to pipeline definitions
  • Configure artifacts for SBOM storage
  • Implement quality gates with vulnerability thresholds

secret-management

  • Integrate secret scanning (Gitleaks, TruffleHog)
  • Automate secret rotation on detection
  • Use pre-commit hooks for prevention

infrastructure-as-code

  • Scan Terraform and Kubernetes manifests with Trivy config
  • Detect misconfigurations before deployment
  • Enforce policy-as-code with OPA

security-hardening

  • Apply remediation guidance from scan results
  • Select secure base images
  • Implement security best practices

compliance-frameworks

  • Generate SBOMs for SOC2, ISO 27001 audits
  • Track vulnerability metrics for compliance reporting
  • Provide evidence for security controls

Quick Reference

Essential Commands

# Trivy: Scan image with severity filter
trivy image --severity HIGH,CRITICAL myapp:latest

# Trivy: Generate SBOM
trivy image --format cyclonedx --output sbom.json myapp:latest

# Trivy: Scan SBOM
trivy sbom sbom.json

# Grype: Scan image
grype myapp:latest --fail-on high

# Syft + Grype: SBOM workflow
syft myapp:latest -o json | grype

# Gitleaks: Scan for secrets
gitleaks detect --source . --verbose

Common Patterns

# CI/CD: Fail build on Critical
trivy image --exit-code 1 --severity CRITICAL myapp:latest

# Ignore unfixed vulnerabilities
trivy image --ignore-unfixed --severity HIGH,CRITICAL myapp:latest

# Scan only OS packages
trivy image --vuln-type os myapp:latest

# Skip specific directories
trivy fs --skip-dirs node_modules,vendor .

Progressive Disclosure

This skill provides foundational vulnerability management patterns. For deeper topics:

  • Tool Selection: references/tool-selection.md - Complete decision frameworks
  • SBOM Patterns: references/sbom-guide.md - Generation, storage, consumption
  • Prioritization: references/prioritization-framework.md - CVSS/EPSS/KEV automation
  • CI/CD Integration: references/ci-cd-patterns.md - GitLab CI, Jenkins, Azure Pipelines
  • Remediation: references/remediation-workflows.md - SLA tracking, false positives
  • Policy-as-Code: references/policy-as-code.md - OPA examples, security gates

Working Examples:

  • examples/trivy/ - Trivy scanning patterns
  • examples/grype/ - Grype + Syft workflows
  • examples/ci-cd/ - Complete pipeline configurations
  • examples/sbom/ - SBOM generation and management
  • examples/prioritization/ - EPSS and KEV integration scripts

Automation Scripts:

  • scripts/vulnerability-report.sh - Generate executive reports
  • scripts/sla-tracker.sh - Track remediation SLAs
  • scripts/false-positive-manager.sh - Manage suppression rules

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenCode

29.96%
按下载量换算57

Gemini CLI

23.74%
按下载量换算45

Antigravity

17.14%
按下载量换算33

Claude Code

14.2%
按下载量换算27

roo

9.01%
按下载量换算17

Cursor

3.7%
按下载量换算7

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。

来源信息

继续浏览同类 Skills