skills/building-incident-response-dashboard/SKILL.md
在 Splunk、Elastic 或 Grafana 中构建实时事件响应(Incident Response)仪表盘,为安全运营中心(SOC)分析师和领导层提供主动事件处理过程中的态势感知(Situational Awareness),追踪受影响系统、遏制状态、失陷指标(IOC)扩散和响应时间线。适用于 IR 团队在事件协调和事后报告期间需要统一可见性的场景。
npx skillsauth add killvxk/cybersecurity-skills-zh building-incident-response-dashboardInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
在以下情况使用本技能:
不适用于日常 SOC 监控仪表盘(应使用事件审查功能)——IR 仪表盘专为主动事件协调和管理报告而设计。
构建 Splunk Dashboard Studio 仪表盘用于主动事件跟踪:
<dashboard version="2" theme="dark">
<label>主动事件响应仪表盘</label>
<description>IR-2024-0450 实时跟踪</description>
<row>
<panel>
<title>事件摘要</title>
<single>
<search>
<query>
| makeresults
| eval incident_id="IR-2024-0450",
status="CONTAINMENT",
severity="Critical",
affected_hosts=7,
contained_hosts=5,
iocs_identified=23,
hours_elapsed=round((now()-strptime("2024-03-15 14:00","%Y-%m-%d %H:%M"))/3600,1)
| table incident_id, status, severity, affected_hosts, contained_hosts, iocs_identified, hours_elapsed
</query>
</search>
</single>
</panel>
</row>
</dashboard>
跟踪受影响系统及其遏制状态:
| inputlookup ir_affected_systems.csv
| eval status_color = case(
status="Contained", "#2ecc71",
status="Compromised", "#e74c3c",
status="Investigating", "#f39c12",
status="Recovered", "#3498db",
1=1, "#95a5a6"
)
| stats count by status
| eval order = case(status="Compromised", 1, status="Investigating", 2,
status="Contained", 3, status="Recovered", 4)
| sort order
| table status, count
--- 详细主机表
| inputlookup ir_affected_systems.csv
| lookup asset_lookup_by_cidr ip AS host_ip OUTPUT category, owner, priority
| table hostname, host_ip, category, owner, status, containment_time,
compromise_vector, analyst_assigned
| sort status, hostname
监控 IOC 在环境中的扩散:
--- 事件期间识别的 IOC
index=* (src_ip IN ("185.234.218.50", "45.77.123.45") OR
dest IN ("evil-c2.com", "malware-drop.com") OR
file_hash IN ("a1b2c3d4...", "e5f6a7b8..."))
earliest="2024-03-14"
| stats count AS hits, dc(src_ip) AS unique_sources,
dc(dest) AS unique_dests, latest(_time) AS last_seen
by sourcetype
| sort - hits
--- IOC 时间线
index=* (src_ip IN ("185.234.218.50") OR dest="evil-c2.com")
earliest="2024-03-14"
| timechart span=1h count by sourcetype
--- 新 IOC 发现跟踪
| inputlookup ir_ioc_list.csv
| stats count by ioc_type, source, discovery_time
| sort discovery_time
| table discovery_time, ioc_type, ioc_value, source, status
创建按时间顺序排列的事件时间线:
| inputlookup ir_timeline.csv
| sort _time
| eval phase = case(
action_type="detection", "检测(Detection)",
action_type="triage", "分诊(Triage)",
action_type="containment", "遏制(Containment)",
action_type="eradication", "根除(Eradication)",
action_type="recovery", "恢复(Recovery)",
1=1, "其他"
)
| eval phase_color = case(
phase="检测(Detection)", "#e74c3c",
phase="分诊(Triage)", "#f39c12",
phase="遏制(Containment)", "#e67e22",
phase="根除(Eradication)", "#2ecc71",
phase="恢复(Recovery)", "#3498db"
)
| table _time, phase, action, analyst, details
时间线数据示例:
_time,action_type,action,analyst,details
2024-03-15 14:00,detection,Alert triggered - Cobalt Strike beacon detected,splunk_es,Notable event NE-2024-08921
2024-03-15 14:12,triage,Alert triaged - confirmed true positive,analyst_jdoe,VT score 52/72 on beacon hash
2024-03-15 14:23,containment,Host WORKSTATION-042 isolated,analyst_jdoe,CrowdStrike network isolation
2024-03-15 14:35,containment,C2 domain blocked on firewall,analyst_msmith,Palo Alto rule deployed
2024-03-15 15:00,eradication,Enterprise-wide IOC scan initiated,analyst_jdoe,Splunk search across all indices
2024-03-15 15:30,containment,3 additional hosts identified and isolated,analyst_msmith,Lateral movement confirmed
2024-03-15 16:00,eradication,Malware removed from all affected hosts,analyst_tier3,CrowdStrike RTR cleanup
2024-03-15 18:00,recovery,Systems restored and monitored,analyst_msmith,72-hour monitoring period started
跟踪整体 SOC 性能指标:
--- 按严重性划分的事件量(过去 30 天)
index=notable earliest=-30d
| stats count by urgency
| eval order = case(urgency="critical", 1, urgency="high", 2, urgency="medium", 3,
urgency="low", 4, urgency="informational", 5)
| sort order
--- 平均检测时间(MTTD)
index=notable earliest=-30d status_label="Resolved*"
| eval mttd_minutes = round((time_of_first_event - orig_time) / 60, 1)
| stats avg(mttd_minutes) AS avg_mttd, median(mttd_minutes) AS med_mttd,
perc95(mttd_minutes) AS p95_mttd
--- 平均响应时间(MTTR)
index=notable earliest=-30d status_label="Resolved*"
| eval mttr_hours = round((status_end - _time) / 3600, 1)
| stats avg(mttr_hours) AS avg_mttr, median(mttr_hours) AS med_mttr by urgency
--- 分析师工作负载分布
index=notable earliest=-7d
| stats count by owner
| sort - count
--- 告警处置分类
index=notable earliest=-30d status_label IN ("Resolved*", "Closed*")
| stats count by disposition
| eval percentage = round(count / sum(count) * 100, 1)
| sort - count
在重大事件期间为领导层创建高级别仪表盘:
--- 高管摘要面板
| makeresults
| eval metrics = "业务影响:1 台文件服务器离线(财务部门),"
."预计恢复:4 小时,"
."数据丢失风险:低(备份已验证),"
."客户影响:无,"
."监管通知:不需要(未确认 PII 暴露)"
--- 趋势对比(本月与上月)
index=notable earliest=-60d
| eval period = if(_time > relative_time(now(), "-30d"), "本月", "上月")
| stats count by period, urgency
| chart sum(count) AS incidents by period, urgency
--- 主要威胁类别
index=notable earliest=-30d
| top rule_name limit=10
| table rule_name, count, percent
使用 Splunk 计划搜索维护仪表盘数据:
--- 计划搜索:更新受影响系统查找表(每 5 分钟运行一次)
index=* (src_ip IN [| inputlookup ir_ioc_list.csv | search ioc_type="ip"
| fields ioc_value | rename ioc_value AS src_ip])
earliest=-1h
| stats latest(_time) AS last_seen, count AS event_count,
values(sourcetype) AS data_sources by src_ip
| eval status = if(last_seen > relative_time(now(), "-15m"), "Active", "Dormant")
| outputlookup ir_affected_systems_auto.csv
| 术语 | 定义 | |------|-----------| | 态势感知(Situational Awareness) | 实时了解事件范围、受影响系统和响应进展 | | MTTD | 平均检测时间(Mean Time to Detect)——从威胁发生到 SOC 告警生成的平均时间 | | MTTR | 平均响应时间(Mean Time to Respond)——从告警到事件解决或遏制的平均时间 | | 遏制率(Containment Rate) | 相对于已被攻陷的系统总数,成功隔离的受影响系统百分比 | | 燃尽图(Burn-Down Chart) | 事件期间随时间推移跟踪剩余未完成调查任务的可视化图表 | | 高管汇报(Executive Briefing) | 显示业务影响、时间线和恢复状态的非技术性摘要仪表盘 |
事件响应仪表盘 — IR-2024-0450
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
状态:遏制阶段(已历时 6 小时 30 分钟)
受影响系统: 遏制进度:
已攻陷: 2 [==========----------] 71%
调查中: 1 7 个系统中已遏制 5 个
已遏制: 3
已恢复: 1
IOC 摘要: 响应时间线:
IP: 4 14:00 — 触发告警
域名: 2 14:12 — 确认恶意
哈希: 3 14:23 — 第一台主机隔离
URL: 5 15:00 — 启动企业扫描
邮件: 1 15:30 — 又隔离 3 台主机
关键指标:
MTTD: 12 分钟
MTTC: 23 分钟(第一台主机)
在职分析师:3 人(Tier 2:2 人,Tier 3:1 人)
业务影响:低 — 财务文件服务器离线,无面向客户的系统受影响
testing
设计并执行社会工程学渗透测试,包括钓鱼、语音钓鱼、短信钓鱼和物理借口活动,以衡量人员安全韧性并识别培训差距。
testing
主持结构化的事件后审查,以识别根本原因、记录有效和无效的措施,并提出可操作的改进建议以提升未来的事件响应能力。
testing
通过分析举报的邮件、提取指标、评估凭据受攻陷情况、在全组织范围隔离恶意邮件并修复受影响账号来响应网络钓鱼事件。涵盖邮件头分析、URL/附件沙箱检测和邮箱范围清除操作。适用于网络钓鱼响应、邮件事件、凭据钓鱼、鱼叉式网络钓鱼调查或钓鱼修复相关请求。
tools
票据传递(Pass-the-Ticket,PtT)是一种横向移动技术,使用窃取的 Kerberos 票据(TGT 或 TGS)在不知道用户密码的情况下向服务进行认证。通过从已控制的主机内存中提取 Kerberos 票据,攻击者可以将这些票据注入自己的会话以模拟票据所有者。