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

building-cloud-security-posture-management构建云安全态势管理

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

285

周安装

12

GitHub Stars

5,908

下载量

1
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:building-cloud-security-posture-management(构建云安全态势管理)
来源仓库:https://github.com/mukul975/anthropic-cybersecurity-skills
仓库路径:skills/building-cloud-security-posture-management
安装命令:
npx skills add https://github.com/mukul975/anthropic-cybersecurity-skills --skill building-cloud-security-posture-management
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/mukul975/anthropic-cybersecurity-skills --skill building-cloud-security-posture-management

简介

用于辅助云环境的安全审计、权限检查和漏洞排查,适合梳理敏感配置与鉴权逻辑。

  • 适用于多账号云环境的态势监控、合规基线建立及迁移前的安全评估场景。
  • 通过分析配置风险生成复核清单,需结合最小权限原则确认操作边界和脱敏方式。
  • 安装使用 GitHub 仓库,涉及密钥或生产系统时应先验证权限与数据保护策略。
  • building-cloud-security-posture-management 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Building Cloud Security Posture Management

When to Use

  • When an organization lacks visibility into cloud misconfigurations across multiple accounts and providers
  • When compliance requirements demand continuous posture monitoring against CIS, NIST, or SOC 2 frameworks
  • When security teams need to prioritize which misconfigurations to remediate based on actual risk
  • When migrating workloads to the cloud and establishing security baselines before production deployment
  • When integrating cloud posture findings into an existing SOC or SIEM platform

Do not use for runtime threat detection (see detecting-cloud-threats-with-guardduty), for application-level vulnerability scanning (see securing-serverless-functions), or for network traffic analysis (see implementing-cloud-network-segmentation).

Prerequisites

  • Cloud accounts across target providers (AWS, Azure, GCP) with read-only API access for CSPM tools
  • Defined compliance framework requirements (CIS Benchmarks, NIST 800-53, PCI-DSS, SOC 2)
  • SIEM or ticketing system for finding ingestion and workflow management
  • Budget allocation for commercial CSPM tooling or engineering capacity for native tool integration

Workflow

Step 1: Assess Current Cloud Estate and Risk Appetite

Inventory all cloud accounts, subscriptions, and projects. Classify them by data sensitivity, regulatory requirements, and business criticality to determine CSPM coverage scope.

Cloud Estate Inventory:
+----------------+----------+------------+--------------------+------------------+
| Provider       | Accounts | Workloads  | Data Classification| Compliance Needs |
+----------------+----------+------------+--------------------+------------------+
| AWS            | 45       | Production | Confidential       | PCI-DSS, SOC 2   |
| AWS            | 12       | Dev/Test   | Internal           | SOC 2            |
| Azure          | 8        | Production | Restricted (PII)   | GDPR, SOC 2      |
| GCP            | 3        | Analytics  | Confidential       | SOC 2            |
+----------------+----------+------------+--------------------+------------------+

Step 2: Select and Deploy CSPM Tooling

Evaluate CSPM solutions based on multi-cloud support, policy coverage, agentless scanning, attack path analysis, and integration capabilities.

Native Tools:

  • AWS Security Hub CSPM with Config rules
  • Microsoft Defender for Cloud CSPM
  • Google Security Command Center Premium

Commercial Platforms:

  • Wiz: Agentless, graph-based visibility, attack path analysis, highest market mindshare (20.2%)
  • Prisma Cloud (now Cortex Cloud): CSPM + CWP + CIEM, 3,000+ built-in policies
  • Orca Security: SideScanning technology, agentless full-stack visibility
  • Lacework: Anomaly-based detection with behavioral analysis
# Example: Deploy Wiz connector for AWS using CloudFormation
aws cloudformation create-stack \
  --stack-name wiz-connector \
  --template-url https://wiz-advanced-security.s3.amazonaws.com/wiz-aws-connector.yaml \
  --parameters ParameterKey=ExternalId,ParameterValue=<wiz-external-id> \
  --capabilities CAPABILITY_NAMED_IAM

# Example: Configure Prisma Cloud AWS onboarding
# Prisma Cloud uses a cross-account IAM role for read-only access
aws iam create-role \
  --role-name PrismaCloudReadOnly \
  --assume-role-policy-document '{
    "Version": "2012-10-17",
    "Statement": [{
      "Effect": "Allow",
      "Principal": {"AWS": "arn:aws:iam::188619942792:root"},
      "Action": "sts:AssumeRole",
      "Condition": {"StringEquals": {"sts:ExternalId": "<prisma-external-id>"}}
    }]
  }'

Step 3: Define Policy Baselines and Custom Rules

Map compliance framework controls to CSPM policies. Create custom rules for organization-specific requirements that go beyond standard benchmarks.

# Example custom CSPM policy definitions
policies:
  - name: s3-bucket-encryption-required
    description: All S3 buckets must have AES-256 or KMS encryption enabled
    provider: aws
    resource_type: aws_s3_bucket
    severity: HIGH
    rule: |
      resource.encryption.rules[0].apply_server_side_encryption_by_default.sse_algorithm
      in ["aws:kms", "AES256"]
    remediation: Enable default encryption on the S3 bucket using AES-256 or AWS KMS
    compliance_mapping:
      - CIS_AWS_v5.0: "2.1.1"
      - PCI_DSS: "3.4"
      - SOC2: "CC6.1"

  - name: public-ip-not-attached-to-compute
    description: Production compute instances must not have public IP addresses
    provider: aws
    resource_type: aws_ec2_instance
    severity: CRITICAL
    rule: |
      resource.public_ip_address == null AND
      resource.tags["Environment"] == "production"
    remediation: Remove public IP and route traffic through a load balancer or NAT gateway

  - name: storage-account-private-endpoint
    description: Azure storage accounts must use private endpoints only
    provider: azure
    resource_type: azurerm_storage_account
    severity: HIGH
    rule: |
      resource.network_rules.default_action == "Deny" AND
      resource.private_endpoint_connections.length > 0

Step 4: Automate Drift Detection and Alerting

Configure continuous scanning intervals, drift detection thresholds, and alert routing to ensure new misconfigurations are detected within minutes of resource creation or modification.

# AWS Config rule for drift detection on S3 public access
aws configservice put-config-rule \
  --config-rule '{
    "ConfigRuleName": "s3-bucket-public-read-prohibited",
    "Source": {
      "Owner": "AWS",
      "SourceIdentifier": "S3_BUCKET_PUBLIC_READ_PROHIBITED"
    },
    "Scope": {"ComplianceResourceTypes": ["AWS::S3::Bucket"]}
  }'

# Auto-remediation using SSM Automation
aws configservice put-remediation-configurations \
  --remediation-configurations '[{
    "ConfigRuleName": "s3-bucket-public-read-prohibited",
    "TargetType": "SSM_DOCUMENT",
    "TargetId": "AWS-DisableS3BucketPublicReadWrite",
    "Automatic": true,
    "MaximumAutomaticAttempts": 3,
    "RetryAttemptSeconds": 60
  }]'

Step 5: Prioritize Findings with Context-Aware Risk Scoring

Move beyond severity-only prioritization. Use attack path analysis, asset context, and exploitability data to focus remediation on findings that represent actual risk.

Risk Prioritization Matrix:
+----------------------------+----------+-----------+--------+-------------+
| Finding                    | Severity | Exposed   | Attack | Priority    |
|                            |          | Internet? | Path?  | Score       |
+----------------------------+----------+-----------+--------+-------------+
| S3 bucket public read      | HIGH     | Yes       | Yes    | CRITICAL    |
| RDS no encryption at rest  | HIGH     | No        | No     | MEDIUM      |
| SG allows 0.0.0.0/0:22    | HIGH     | Yes       | Yes    | CRITICAL    |
| CloudTrail not enabled     | MEDIUM   | No        | No     | HIGH        |
| EBS volume not encrypted   | MEDIUM   | No        | No     | LOW         |
+----------------------------+----------+-----------+--------+-------------+

Step 6: Integrate with SOC Workflows and Reporting

Feed CSPM findings into SIEM platforms, create Jira tickets for remediation tracking, and build executive dashboards for posture trending.

# Export findings to Amazon Security Lake in OCSF format
aws securitylake create-subscriber \
  --subscriber-name cspm-siem-integration \
  --sources '[{"awsLogSource": {"sourceName": "SH_FINDINGS"}}]' \
  --subscriber-identity '{"principal": "arn:aws:iam::123456789012:role/SIEMIngestionRole", "externalId": "siem-ext-id"}'

Key Concepts

TermDefinition
CSPMCloud Security Posture Management: continuous monitoring service that identifies cloud infrastructure misconfigurations and compliance violations
Configuration DriftDeviation from a defined security baseline that occurs when resources are modified outside of approved change management processes
Attack PathA multi-step chain of misconfigurations and vulnerabilities that an adversary could exploit to move from an entry point to a critical asset
Agentless ScanningCSPM approach that uses cloud provider APIs and snapshot analysis to assess security posture without installing agents on workloads
Policy as CodeDefining security policies in machine-readable formats (Rego, YAML, JSON) that can be version-controlled and automatically enforced
Compliance FrameworkStructured set of security controls and requirements such as CIS Benchmarks, NIST 800-53, PCI-DSS, or SOC 2 used to measure posture
Security GraphGraph database representing relationships between cloud resources, identities, network paths, and vulnerabilities for contextual risk analysis

Tools & Systems

  • Wiz: Agentless CNAPP providing graph-based CSPM, attack path analysis, and vulnerability management across all major cloud providers
  • Prisma Cloud / Cortex Cloud: Palo Alto Networks CNAPP with 3,000+ built-in policies covering CSPM, CWP, CIEM, and IaC security
  • AWS Security Hub CSPM: Native AWS posture management with automated checks against CIS v5.0 and AWS Foundational Security Best Practices
  • Prowler: Open-source AWS/Azure/GCP security assessment tool with 300+ checks and CIS benchmark support
  • Steampipe: Open-source SQL-based cloud configuration querying tool supporting 140+ plugins for multi-cloud posture queries

Common Scenarios

Scenario: Post-Acquisition Cloud Posture Assessment

Context: A company acquires a startup with 30 AWS accounts and 5 GCP projects. No CSPM tooling is in place and the security team needs to assess the inherited environment within two weeks.

Approach:

  1. Deploy an agentless CSPM tool (Wiz or Orca) using read-only cross-account roles for immediate visibility without agent installation
  2. Run initial scans against CIS Benchmarks for both AWS and GCP to establish a baseline posture score
  3. Identify Critical findings: publicly exposed databases, unencrypted storage with sensitive data, overprivileged service accounts
  4. Prioritize attack paths that connect internet-exposed resources to data stores containing customer PII
  5. Deliver an executive summary with risk-ranked findings and a 90-day remediation roadmap
  6. Integrate the acquired accounts into the existing CSPM platform with continuous monitoring

Pitfalls: Deploying agents for the initial assessment adds weeks of delay. Using only native tools for a multi-cloud assessment creates separate dashboards and makes cross-cloud comparison difficult.

Output Format

Cloud Security Posture Assessment Report
==========================================
Organization: Acme Corp
Cloud Providers: AWS (57 accounts), Azure (8 subscriptions), GCP (3 projects)
CSPM Platform: Wiz
Assessment Date: 2025-02-23

OVERALL POSTURE SCORE: 68/100

FINDINGS BY SEVERITY:
  Critical: 47   (Internet-exposed + data access risk)
  High: 234      (Misconfiguration with limited exposure)
  Medium: 891    (Non-compliant but low immediate risk)
  Low: 1,567     (Informational or best practice)

TOP ATTACK PATHS:
  1. Internet -> Public S3 Bucket (PII data) -> No encryption
     Affected: 3 accounts | Risk: Critical | ETA to remediate: 1 day
  2. Internet -> EC2 (SSH open) -> IAM Role -> Cross-Account Admin
     Affected: 1 account | Risk: Critical | ETA to remediate: 2 days
  3. Internet -> Azure App Service -> SQL Server (public endpoint)
     Affected: 2 subscriptions | Risk: Critical | ETA to remediate: 3 days

COMPLIANCE STATUS:
  CIS AWS v5.0:         62% compliant (340/548 controls passing)
  CIS Azure v4.0:       71% compliant (189/266 controls passing)
  CIS GCP v4.0:         58% compliant (87/150 controls passing)
  SOC 2 Type II:        74% controls mapped and passing

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

33.93%
按下载量换算0

Claude

32.75%
按下载量换算0

Cursor

19.43%
按下载量换算0

Gemini CLI

9.54%
按下载量换算0

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