Token导航 LogoToken导航TokenDH.com
研究检索敏感数据github未标认证来源可访问clear审计通过

cloud-collector云收集器

Agent Skill

cloud-collector 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

196

周安装

8

GitHub Stars

公开资料未说明

下载量

63
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:cloud-collector(云收集器)
来源仓库:https://github.com/workstreet-compliance/cloud-collector
仓库路径:skills/cloud-collector
安装命令:
npx skills add https://github.com/workstreet-compliance/cloud-collector --skill cloud-collector
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/workstreet-compliance/cloud-collector --skill cloud-collector

简介

自动化整理 AWS/GCP/Azure 环境的合规证据包。

  • 支持 SOC 2、ISO 27001 与 CIS 基准的配置验证。
  • 生成审计报告所需的安全控制项检查结果摘要。cloud-collector 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 需预先配置各云平台有效凭证方可启动收集作业
  • 输出结果仅供审计参考,不得替代人工安全评估

SKILL.md

Evidence Collector Skill

Automated compliance evidence collection for AWS, GCP, and Azure cloud environments.

When to Activate

Activate this skill when the user:

  • Asks to collect compliance evidence
  • Mentions SOC 2, ISO 27001, NIST 800-53, or CIS benchmarks
  • Requests audit documentation or evidence packages
  • Wants to verify security configurations across cloud providers
  • Asks about IAM policies, logging, encryption, or network security evidence

Prerequisites

Before running evidence collection, ensure:

  1. AWS: Valid AWS credentials configured via:

- Environment variables (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY) - AWS credentials file (~/.aws/credentials) - IAM role (if running on EC2/Lambda) - Required permissions: iam:Get*, iam:List*, cloudtrail:Describe*, cloudtrail:Get*, s3:GetBucket*, s3:ListBucket, securityhub:Get*, kms:Describe*, kms:List*, ec2:Describe*

  1. GCP: Valid GCP credentials configured via:

- Application Default Credentials (gcloud auth application-default login) - Service account key file (GOOGLE_APPLICATION_CREDENTIALS) - Required roles: roles/iam.securityReviewer, roles/logging.viewer, roles/storage.objectViewer, roles/securitycenter.findingsViewer, roles/cloudkms.viewer, roles/compute.viewer

  1. Azure: Valid Azure credentials configured via:

- Azure CLI (az login) - Service principal environment variables - Required roles: Reader, Security Reader, Key Vault Reader

Usage Patterns

Collect All Evidence for a Cloud Provider

from scripts.aws_evidence import AWSEvidenceCollector
from scripts.output_formatter import EvidenceFormatter

# Collect AWS evidence
collector = AWSEvidenceCollector()
package = collector.collect_all()

# Output as JSON
print(EvidenceFormatter.to_json(package))

# Output as Markdown report
print(EvidenceFormatter.to_markdown(package))

# Save to files
EvidenceFormatter.save(package, "./evidence_output")

Collect Specific Evidence Categories

from scripts.aws_evidence import AWSEvidenceCollector

collector = AWSEvidenceCollector()

# Collect only IAM evidence
iam_evidence = collector.collect_iam()

# Collect only logging evidence
logging_evidence = collector.collect_cloudtrail()

# Collect encryption evidence
encryption_evidence = collector.collect_kms()

Multi-Cloud Collection

from scripts.aws_evidence import AWSEvidenceCollector
from scripts.gcp_evidence import GCPEvidenceCollector
from scripts.azure_evidence import AzureEvidenceCollector
from scripts.output_formatter import EvidenceFormatter

# Collect from all providers
aws_package = AWSEvidenceCollector().collect_all()
gcp_package = GCPEvidenceCollector(project_id="my-project").collect_all()
azure_package = AzureEvidenceCollector(subscription_id="sub-id").collect_all()

# Save all packages
for package in [aws_package, gcp_package, azure_package]:
    EvidenceFormatter.save(package, "./evidence_output")

Evidence Categories

Each cloud provider collector gathers evidence in these categories:

CategoryDescriptionControls Supported
IAMIdentity policies, roles, users, groupsCC6.1, CC6.2, CC6.3, A.9.2, AC-2, AC-3
LoggingAudit trails, log configurationsCC7.2, A.12.4, AU-2, AU-3, AU-12
StorageBucket/blob policies, access controlsCC6.1, A.8.2, AC-3, SC-28
SecuritySecurity findings, vulnerabilitiesCC7.1, A.12.6, SI-4, RA-5
EncryptionKey management, encryption configsCC6.1, A.10.1, SC-12, SC-13
NetworkFirewall rules, security groups, NSGsCC6.6, A.13.1, SC-7, AC-4

Output Formats

JSON Output

Structured JSON following the schema in references/evidence_schema.json. Suitable for:

  • Automated processing
  • Integration with GRC tools
  • Long-term evidence storage

Markdown Output

Human-readable report with:

  • Metadata summary
  • Evidence grouped by category
  • Control mapping tables with status indicators
  • Suitable for auditor review and documentation

Control Framework Mappings

Evidence is automatically mapped to controls from:

  • SOC 2 Type II: CC (Common Criteria) series
  • ISO 27001: Annex A controls
  • NIST 800-53: Security and privacy controls
  • CIS Benchmarks: Cloud-specific benchmarks

See references/control_mappings.md for complete mapping details.

Error Handling

The collectors handle common errors gracefully:

  • Missing Permissions: Logs warning, continues with available data
  • API Rate Limits: Implements exponential backoff
  • Region Unavailable: Skips region, notes in output
  • Resource Not Found: Records as "not configured" evidence

Best Practices

  1. Run with least privilege: Use read-only credentials
  2. Scope appropriately: Collect only what you need for the audit
  3. Timestamp everything: Evidence is timestamped automatically
  4. Version control outputs: Store evidence packages in version control
  5. Review before submission: Always review collected evidence before sharing with auditors

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

Claude Code

29.56%
按下载量换算19

Codex

24.34%
按下载量换算15

windsurf

16.29%
按下载量换算10

trae

11.33%
按下载量换算7

OpenCode

6.95%
按下载量换算4

Cursor

3.62%
按下载量换算2

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

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。

来源信息

继续浏览同类 Skills