external/anthropic-cybersecurity-skills/skills/implementing-soar-playbook-with-palo-alto-xsoar/SKILL.md
Implement automated incident response playbooks in Cortex XSOAR to orchestrate security workflows across SOC tools and reduce manual response time.
npx skillsauth add seikaikyo/dash-skills implementing-soar-playbook-with-palo-alto-xsoarInstall 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.
Cortex XSOAR (formerly Demisto) is Palo Alto Networks' Security Orchestration, Automation, and Response platform. Playbooks are the core automation engine in XSOAR, enabling SOC teams to automate repetitive incident response tasks. XSOAR provides 900+ prebuilt integration packs, 87 common playbooks, and a visual drag-and-drop editor for building custom workflows. Organizations using SOAR automation reduce mean time to respond (MTTR) by 80% on average.
Incident Type (e.g., Phishing)
|
v
Incident Layout (UI display configuration)
|
v
Pre-Processing Rules (auto-classification, deduplication)
|
v
Playbook (automation logic)
|-- Sub-Playbooks (modular reusable workflows)
|-- Tasks (individual automation steps)
|-- Conditional Tasks (decision branches)
|-- Scripts (custom Python/JavaScript)
|-- Integrations (external tool commands)
|
v
War Room (investigation timeline)
|
v
Closing Report
| Task Type | Purpose | Example |
|---|---|---|
| Standard | Execute a command | !ip ip=8.8.8.8 |
| Conditional | Branch logic | If severity > high, escalate |
| Manual | Require analyst input | Approve containment action |
| Section Header | Organize workflow | "Enrichment Phase" |
| Data Collection | Gather external data | Ask user for additional details |
| Timer | Wait for condition/time | Wait 5 minutes then check |
incident_type: Phishing
playbook: Phishing Investigation - Full
severity_mapping:
- condition: email contains executable attachment
severity: high
- condition: email from external domain with link
severity: medium
- condition: email reported by user
severity: low
layout: Phishing Layout
sla: 60 minutes
id: phishing-investigation-full
version: -1
name: Phishing Investigation - Full
description: Automated phishing email investigation with enrichment, analysis, and response
starttaskid: "0"
tasks:
"0":
id: "0"
taskid: start
type: start
nexttasks:
'#none#':
- "1"
"1":
id: "1"
taskid: extract-indicators
type: regular
task:
name: Extract Indicators from Email
script: ParseEmailFiles
nexttasks:
'#none#':
- "2"
- "3"
- "4"
"2":
id: "2"
taskid: enrich-urls
type: playbook
task:
name: URL Enrichment
playbookName: URL Enrichment - Generic v2
"3":
id: "3"
taskid: enrich-files
type: playbook
task:
name: File Enrichment
playbookName: File Enrichment - Generic v2
"4":
id: "4"
taskid: enrich-ips
type: playbook
task:
name: IP Enrichment
playbookName: IP Enrichment - Generic v2
"5":
id: "5"
taskid: determine-verdict
type: condition
task:
name: Is Email Malicious?
conditions:
- label: "yes"
condition:
- - operator: isEqualString
left: DBotScore.Score
right: "3"
- label: "no"
nexttasks:
"yes":
- "6"
"no":
- "9"
"6":
id: "6"
taskid: block-sender
type: regular
task:
name: Block Sender Domain
script: '|||o365-mail-block-sender'
scriptarguments:
sender_address: ${incident.emailfrom}
"7":
id: "7"
taskid: search-mailboxes
type: regular
task:
name: Search and Delete from All Mailboxes
script: '|||o365-mail-purge-compliance-search'
scriptarguments:
query: "from:${incident.emailfrom} subject:${incident.emailsubject}"
"8":
id: "8"
taskid: notify-user
type: regular
task:
name: Notify Reporting User
script: '|||send-mail'
scriptarguments:
to: ${incident.reporter}
subject: "Phishing Report Confirmed - Action Taken"
body: "The email you reported has been confirmed as malicious and removed."
"9":
id: "9"
taskid: close-incident
type: regular
task:
name: Close Incident
script: closeInvestigation
!ParseEmailFiles entryid=${File.EntryID}
!rasterize url=${URL.Data} type=png
!url url=${URL.Data}
!file file=${File.SHA256}
!ip ip=${IP.Address}
!domain domain=${Domain.Name}
!o365-mail-block-sender sender=${incident.emailfrom}
!o365-mail-purge-compliance-search query="from:${incident.emailfrom}"
!pan-os-block-ip ip=${IP.Address} log_forwarding="default"
!cortex-xdr-isolate-endpoint endpoint_id=${Endpoint.ID}
!jira-create-issue summary="Phishing Incident - ${incident.id}" type="Incident" priority="High"
!servicenow-create-ticket short_description="Security Incident" urgency="2"
Trigger: Malware alert from EDR
Steps:
1. Extract file hash, process details, host info
2. Enrich hash via VirusTotal, Hybrid Analysis
3. Check if file is on allowlist
4. If malicious:
a. Isolate endpoint via EDR
b. Block hash on all endpoints
c. Search for hash across environment
d. Create incident ticket
5. If clean: Close as false positive
Trigger: Impossible travel or suspicious login alert
Steps:
1. Get user details from Active Directory
2. Get login history for past 30 days
3. Check for impossible travel (geo-distance vs time)
4. Check for known VPN/proxy IP
5. If compromised:
a. Disable AD account
b. Revoke all OAuth tokens
c. Reset MFA
d. Notify user's manager
e. Search for lateral movement
6. If false positive: Document and close
Trigger: Network anomaly alert
Steps:
1. Verify traffic spike from network monitoring
2. Identify source IPs and geolocation
3. Check if source IPs are known botnets
4. Implement rate limiting on WAF
5. If sustained attack:
a. Enable upstream DDoS protection
b. Activate CDN scrubbing
c. Notify ISP if needed
6. Monitor and document
# XSOAR Automation Script: CalculateRiskScore
def calculate_risk_score():
"""Calculate composite risk score for an incident."""
severity = demisto.incident().get('severity', 0)
indicator_count = len(demisto.get(demisto.context(), 'DBotScore', []))
malicious_count = len([
i for i in demisto.get(demisto.context(), 'DBotScore', [])
if i.get('Score', 0) == 3
])
base_score = severity * 20
indicator_boost = min(indicator_count * 5, 25)
malicious_boost = malicious_count * 15
risk_score = min(100, base_score + indicator_boost + malicious_boost)
return_results(CommandResults(
outputs_prefix='RiskScore',
outputs={'Score': risk_score, 'Level': 'Critical' if risk_score > 80 else 'High' if risk_score > 60 else 'Medium'},
readable_output=f'Risk Score: {risk_score}/100'
))
calculate_risk_score()
| Metric | Before SOAR | After SOAR | Improvement | |---|---|---|---| | Phishing MTTR | 45 min | 5 min | 89% reduction | | Malware MTTR | 60 min | 8 min | 87% reduction | | Account Compromise MTTR | 30 min | 4 min | 87% reduction | | Alerts Handled per Shift | 50 | 200+ | 300% increase | | False Positive Handling | 10 min | 30 sec | 95% reduction |
development
拋棄式 HTML mockup 比稿:產出 2 到 3 個設計立場不同的變體(密度 / 版式 / 強調軸,不是換色),各附取捨說明,最後給有立場的對比結論。適用:「畫個草圖」「比較 A 版 B 版」「先看方向再做」「給我看幾種做法」。要 production 元件或設計已定案時不適用。
tools
需求不明時的意圖萃取訪談:一次一題、每題附上自己的猜測、聽出「真正想要 vs 覺得應該要」,直到能預測使用者反應(約 95% 信心)才動工。適用:需求缺少對象 / 動機 / 成功標準 / 約束,或使用者點名「訪談我」「先確認一下」「我們確定嗎」。明確自足的指示、純資訊查詢、機械性操作不適用。
development
對非平凡決策啟動新鮮 context 對抗審查(找碴不背書),在修正還便宜的時候抓出錯誤方向。適用:高風險改動(production、資安敏感邏輯、不可逆操作)、不熟的程式碼、要宣稱「這樣是安全的 / 可行的」之前。機械性操作與一行修改不適用。
testing
Reference for writing and editing agent skills well — the vocabulary and principles that make a skill predictable. Consult when authoring, reviewing, or pruning a SKILL.md.