Token导航 LogoToken导航

Agent Skill · 研究检索

collecting-indicators-of-compromise(收集妥协指标)

collecting-indicators-of-compromise 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。

collecting-indicators-of-compromise 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 56,650。

总安装 447周安装 19GitHub Stars 5,892下载 157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:collecting-indicators-of-compromise(收集妥协指标)
来源仓库:https://github.com/mukul975/anthropic-cybersecurity-skills
仓库路径:skills/collecting-indicators-of-compromise
安装命令:
npx skills add https://github.com/mukul975/anthropic-cybersecurity-skills --skill collecting-indicators-of-compromise
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 collecting-indicators-of-compromise

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。

collecting-indicators-of-compromise 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果。适用于事件响应期间识别恶意基础设施、事后取证归档或向 ISAC 共享威胁情报的场景。

支持 IOC 富化与风险评分,协助构建 SIEM/EDR 检测规则,提升防御有效性。需具备 incident response plan 和 forensic toolkit 访问权限方可正常使用。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/collecting-indicators-of-compromise

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/mukul975/anthropic-cybersecurity-skills --skil…;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

collecting-indicators-of-compromise 怎么安装?

在终端执行 npx skills add https://github.com/mukul975/anthropic-cybersecurity-skills --skill collecti…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

collecting-indicators-of-compromise 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。

collecting-indicators-of-compromise 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。

collecting-indicators-of-compromise 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。

collecting-indicators-of-compromise 用的人多吗?

公开数据是累计安装 447 次,近一周安装 19 次,下载 157 次,GitHub 5,892 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。

collecting-indicators-of-compromise 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 5,892 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
collecting-indicators-of-compromise

收集妥协指标 · 本页

4475,892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

Collecting Indicators of Compromise

When to Use

  • During active incident response to identify and block adversary infrastructure
  • Post-incident to document all observed adversary artifacts for future detection
  • When sharing threat intelligence with ISACs, sector partners, or law enforcement
  • When building detection rules in SIEM, EDR, or network security tools
  • When enriching IOCs with threat intelligence context for risk scoring

Do not use for behavioral TTP analysis without accompanying technical indicators; use MITRE ATT&CK mapping for behavioral characterization.

Prerequisites

  • Access to incident evidence sources: SIEM logs, EDR telemetry, memory dumps, disk images, network captures
  • Threat intelligence platform (MISP, OpenCTI, ThreatConnect) for IOC management and sharing
  • IOC enrichment tools: VirusTotal, OTX (AlienVault Open Threat Exchange), Shodan, DomainTools
  • STIX 2.1 knowledge for structured IOC representation
  • Sharing agreements with relevant ISACs (FS-ISAC, H-ISAC, IT-ISAC) or sector partners

Workflow

Step 1: Identify IOC Categories

Collect indicators across all categories from incident evidence:

Network Indicators:

  • IP addresses (C2 servers, staging servers, exfiltration destinations)
  • Domain names (C2 domains, phishing domains, DGA domains)
  • URLs (malware download, C2 check-in, exfiltration endpoints)
  • JA3/JA3S hashes (TLS client/server fingerprints)
  • User-Agent strings (custom or unusual HTTP headers)
  • DNS query patterns (tunneling signatures, DGA patterns)

Host Indicators:

  • File hashes (MD5, SHA-1, SHA-256 of malware, tools, scripts)
  • File paths (known malware installation directories)
  • Registry keys (persistence mechanisms, configuration storage)
  • Scheduled tasks and service names (persistence)
  • Mutex/event names (malware instance synchronization)
  • Named pipes (C2 communication channels, e.g., Cobalt Strike)

Email Indicators:

  • Sender addresses and domains (spoofed or attacker-controlled)
  • Subject lines and body content patterns
  • Attachment names and hashes
  • Embedded URLs
  • Email header anomalies (SPF/DKIM/DMARC failures)

Step 2: Extract IOCs from Evidence Sources

Systematically extract indicators from each evidence source:

From SIEM/Log Analysis:

# Extract unique destination IPs from firewall logs
index=firewall action=blocked
| stats count by dest_ip
| where count > 100

# Extract domains from DNS query logs
index=dns query=*evil* OR query=*c2*
| stats count by query

From Memory Forensics:

# Extract network connections
vol -f memory.raw windows.netscan | grep ESTABLISHED

# Extract strings from suspicious process memory
vol -f memory.raw windows.memmap --pid 3847 --dump
strings -n 8 pid.3847.dmp | grep -E "(http|https)://"

From Malware Analysis:

Sandbox Report IOC Extraction:
- Dropped files:      3 (hashes extracted)
- DNS queries:        update.evil[.]com, cdn.malware[.]net
- HTTP connections:   POST to https://185.220.101[.]42/gate.php
- Registry modified:  HKCU\Software\Microsoft\Windows\CurrentVersion\Run\svcupdate
- Mutex created:      Global\MTX_0x1234ABCD
- Named pipe:         \\.\pipe\MSSE-1234-server

Step 3: Enrich IOCs with Context

Add threat intelligence context to each indicator:

IOC Enrichment Report:
━━━━━━━━━━━━━━━━━━━━━
IP: 185.220.101.42
  VirusTotal:     12/89 vendors flag as malicious
  Shodan:         Open ports: 443, 8443, 80
  Geolocation:    Netherlands, AS208476
  First Seen:     2025-10-01
  Threat Intel:   Associated with Qakbot C2 infrastructure
  Confidence:     High
  TLP:            AMBER

Domain: update.evil[.]com
  Registration:   2025-10-28 (recently registered)
  Registrar:      Namecheap
  WHOIS Privacy:  Yes
  VirusTotal:     8/89 vendors flag as malicious
  DNS History:    Resolved to 185.220.101.42, 91.215.85.17
  Confidence:     High
  TLP:            AMBER

Step 4: Score and Prioritize IOCs

Assign confidence and risk scores to each indicator:

ScoreConfidence LevelCriteria
90-100Confirmed MaliciousMultiple TI sources confirm, observed in active attack
70-89Highly SuspiciousSingle TI source confirms, behavioral analysis supports
50-69SuspiciousLimited TI data, contextually suspicious
30-49UnconfirmedNo TI matches, but anomalous in environment
0-29Likely BenignFalse positive indicators or legitimate infrastructure

Step 5: Distribute IOCs for Detection and Blocking

Push IOCs to defensive systems for immediate protection:

  • Firewall/IPS: Block C2 IPs and domains
  • DNS: Sinkhole malicious domains
  • EDR: Add file hashes to blocklist, create custom IOC watchlists
  • Email Gateway: Block sender domains, attachment hashes, malicious URLs
  • SIEM: Create correlation searches for IOC matches
  • Web Proxy: Block URLs and domains in web filtering policy

Step 6: Share IOCs with Partners

Package IOCs in STIX 2.1 format for sharing:

{
  "type": "indicator",
  "spec_version": "2.1",
  "id": "indicator--a1b2c3d4-e5f6-7890-abcd-ef1234567890",
  "created": "2025-11-15T18:00:00Z",
  "modified": "2025-11-15T18:00:00Z",
  "name": "Qakbot C2 Server IP",
  "indicator_types": ["malicious-activity"],
  "pattern": "[ipv4-addr:value = '185.220.101.42']",
  "pattern_type": "stix",
  "valid_from": "2025-11-15T14:23:00Z",
  "confidence": 95,
  "labels": ["c2", "qakbot"],
  "object_marking_refs": ["marking-definition--f88d31f6-486f-44da-b317-01333bde0b82"]
}

Submit to MISP, ISAC portals, and TAXII servers per sharing agreements.

Key Concepts

TermDefinition
IOC (Indicator of Compromise)Technical artifact observed during a security incident that indicates adversary presence (hash, IP, domain, etc.)
TLP (Traffic Light Protocol)Standard for classifying the sharing restrictions of threat intelligence: WHITE, GREEN, AMBER, AMBER+STRICT, RED
STIX (Structured Threat Information Expression)Standard language for representing cyber threat intelligence in a structured, machine-readable format
TAXII (Trusted Automated Exchange of Intelligence Information)Transport protocol for sharing STIX-formatted threat intelligence between organizations
Confidence ScoreNumerical rating (0-100) indicating the analyst's certainty that an indicator is truly malicious
IOC LifecycleProcess of creating, validating, distributing, and eventually retiring indicators as they lose relevance
DefangingPractice of modifying malicious URLs and domains in reports to prevent accidental clicks (e.g., evil[.]com)

Tools & Systems

  • MISP: Open-source threat intelligence sharing platform for managing, storing, and distributing IOCs
  • VirusTotal: Multi-engine malware scanning and threat intelligence platform for IOC enrichment
  • OpenCTI: Open-source cyber threat intelligence platform supporting STIX 2.1 natively
  • Yeti: Open-source platform for organizing observables, indicators, and TTPs
  • CyberChef: GCHQ's data transformation tool useful for decoding, defanging, and formatting IOCs

Common Scenarios

Scenario: Post-Incident IOC Package for ISAC Sharing

Context: After responding to a Qakbot infection that led to Cobalt Strike deployment, the IR team must package all IOCs for sharing with the Financial Services ISAC (FS-ISAC).

Approach:

  1. Compile all network, host, and email indicators from the investigation
  2. Enrich each IOC with VirusTotal and MISP correlation data
  3. Assign confidence scores based on direct observation vs. secondary correlation
  4. Mark all IOCs with TLP:AMBER for partner sharing
  5. Export as STIX 2.1 bundle and submit to FS-ISAC TAXII feed
  6. Create a human-readable IOC summary report for email distribution

Pitfalls:

  • Including internal IP addresses or hostnames in shared IOC packages (information leakage)
  • Sharing IOCs at TLP:WHITE that should be restricted to TLP:AMBER
  • Not defanging URLs and domains in human-readable reports
  • Sharing IP addresses of legitimate CDNs or cloud providers as malicious IOCs

Output Format

INDICATOR OF COMPROMISE REPORT
================================
Incident:     INC-2025-1547
Date:         2025-11-15
TLP:          AMBER
Sharing:      FS-ISAC, internal SOC

NETWORK INDICATORS
Type     | Value                    | Confidence | Context
---------|--------------------------|------------|--------
IPv4     | 185.220.101[.]42         | 95         | Qakbot C2 server
IPv4     | 91.215.85[.]17           | 90         | Cobalt Strike C2
Domain   | update.evil[.]com        | 95         | Staging domain
URL      | hxxps://185.220[.]101.42/gate.php | 95  | C2 check-in
JA3      | a0e9f5d64349fb13191bc7...| 80         | Qakbot TLS fingerprint

HOST INDICATORS
Type     | Value                    | Confidence | Context
---------|--------------------------|------------|--------
SHA-256  | a1b2c3d4e5f6...         | 100        | Qakbot dropper
SHA-256  | b2c3d4e5f6a7...         | 100        | Cobalt Strike beacon
FilePath | C:\Users\*\AppData\Local\Temp\update.exe | 85 | Dropper location
RegKey   | HKCU\...\Run\svcupdate  | 90         | Persistence
Mutex    | Global\MTX_0x1234ABCD   | 95         | Qakbot instance lock
Task     | WindowsUpdate           | 90         | Scheduled task persistence

EMAIL INDICATORS
Type     | Value                    | Confidence | Context
---------|--------------------------|------------|--------
Sender   | billing@spoofed[.]com   | 95         | Phishing sender
Subject  | "Invoice-Nov2025"       | 70         | Phishing subject line
Hash     | c3d4e5f6a7b8...         | 100        | Malicious .docm attachment

TOTAL: 14 indicators | HIGH confidence avg: 91

平台分布

Codex

37.46%
按下载量换算59

Claude

31.64%
按下载量换算50

Cursor

18.16%
按下载量换算29

Gemini CLI

9.11%
按下载量换算14

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