Menu
Generate demo log data for Splunk. Use when you need to create synthetic logs for testing, demos, or training.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Analyze a log sample to produce a draft SPEC.yaml for adding a new data source. Use when you have raw log lines from a new source and want a structured format analysis before scaffolding a generator.
Create a new log generator following project patterns. Use when adding support for a new log source like Palo Alto, CrowdStrike, Okta, etc.
Create a new attack or operational scenario. Use when adding coordinated events across multiple log sources.
Use this skill when the user wants to create, edit, or generate Splunk Dashboard Studio dashboards (JSON-based, version 2). Triggers include any mention of "Dashboard Studio", "Splunk dashboard", "splunk.pie", "splunk.table", "splunk.singlevalue", "splunk.timeline", "splunk.parallelcoordinates", or requests to build Splunk visualizations, security dashboards, monitoring dashboards, or SOC dashboards. Also use when converting Simple XML dashboards to Dashboard Studio format, or when generating dashboard JSON definitions programmatically. Do NOT use for Classic Simple XML dashboards (version 1), Splunk Observability Cloud dashboards, or AppDynamics dashboards.
Complete field reference for all FAKE: sourcetypes in the fake_tshrt index. Use when writing SPL queries, building dashboards, or understanding the data model.
Create a new REST endpoint for the Splunk TA. Use when adding new functionality to the admin dashboard.
Baseado na classificação ocupacional SOC
| name | generate-logs |
| description | Generate demo log data for Splunk. Use when you need to create synthetic logs for testing, demos, or training. |
| allowed-tools | Bash(python3 *) |
| metadata | {"argument-hint":"[--sources=X,Y] [--days=N] [--scenarios=TYPE]"} |
Generate realistic synthetic log data using the TA-FAKE-TSHRT generators.
By default, generators write to bin/output/tmp/ (test mode, safe — Splunk does NOT monitor this). Use --no-test to write to bin/output/ which Splunk's inputs.conf ingests.
--test # Default. Writes to output/tmp/
--no-test # Production. Writes to output/ (Splunk-monitored)
cd TheFakeTshirtCompany/TA-FAKE-TSHRT/bin
# All sources, test mode (default)
python3 main_generate.py --all --scenarios=all --days=14
# Production run for Splunk ingestion
python3 main_generate.py --all --scenarios=all --days=14 --no-test
# Specific sources with one scenario
python3 main_generate.py --sources=asa,entraid,aws --scenarios=exfil
# High-volume demo (31 days)
python3 main_generate.py --all --days=31 --orders-per-day=3000 --clients=40 --full-metrics --no-test
# Minimal smoke test
python3 main_generate.py --sources=asa --days=1 --quiet
| Category | Sources |
|---|---|
| Network | asa, meraki, catalyst, aci |
| Cloud/Security | aws, aws_guardduty, aws_billing, gcp, entraid, secure_access, catalyst_center |
| Collaboration | exchange, office_audit, webex_ta, webex_api |
| Windows | wineventlog, perfmon, sysmon, mssql |
| Linux | linux |
| Web/Retail | access, orders, servicebus |
| ERP/ITSM | sap, servicenow |
| OT/ICS | cybervision (Detroit plant, 21 OT assets) |
| Group | Expands to |
|---|---|
all | All 26 sources |
cloud | aws, aws_guardduty, aws_billing, gcp, entraid, secure_access |
network | asa, meraki, catalyst, aci, cybervision |
cisco | asa, meraki, secure_access, catalyst, aci, catalyst_center, cybervision |
campus | catalyst, catalyst_center |
datacenter | aci |
ot | cybervision |
windows | wineventlog, perfmon, mssql, sysmon |
linux | linux |
web | access |
office | office_audit, exchange |
email | exchange |
retail | orders, servicebus |
collaboration | webex_ta, webex_api |
erp | sap |
itsm | servicenow |
| Scenario | Category | Days | Target |
|---|---|---|---|
exfil | attack | 1-14 | Alex Miller (Finance, Boston) |
ransomware_attempt | attack | 8-9 | Brooklyn White (Austin) |
phishing_test | attack | 21-23 | All employees (IT campaign) |
ot_rogue_device | attack | 8 | PLC-PRINT-01 (Detroit Zone-1) |
memory_leak | ops | 7-10 | WEB-01 |
cpu_runaway | ops | 11-12 | SQL-PROD-01 |
disk_filling | ops | 1-5 | MON-ATL-01 |
dead_letter_pricing | ops | 16 | WEB-01 (ServiceBus) |
ddos_attack | network | 18-19 | WEB-01 |
firewall_misconfig | network | 6 | FW-EDGE-01 |
certificate_expiry | network | 13 | FW-EDGE-01 |
Meta-values: none, all, attack, ops, network.
--start-date=YYYY-MM-DD Start date (default: 2026-01-01)
--days=N Number of days (default: 14)
--scale=N.N Volume multiplier (default: 1.0)
--scenarios=X,Y Scenarios or category (default: none)
--parallel=N Worker threads (default: 4)
--quiet Suppress progress output
--test / --no-test Output mode (default: --test)
# Perfmon-specific
--clients=N Client workstations (default: 5, min: 5, max: 195)
--client-interval=N Minutes between non-scenario client metrics (default: 30)
--full-metrics Include disk/network for clients (increases volume)
# Orders-specific
--orders-per-day=N Target daily orders (default: ~224)
# Meraki-specific
--meraki-health-interval Health metric frequency (5/10/15/30 min)
Note: scale is ignored by perfmon (fixed intervals), orders (use --orders-per-day), and servicebus (1:1 with orders).
bin/output/tmp/
├── network/ # cisco_asa, meraki_*, catalyst, aci
├── cloud/ # aws, gcp, entraid, exchange, webex, secure_access, catalyst_center
├── windows/ # perfmon, wineventlog, sysmon, mssql
├── linux/ # vmstat, df, iostat, cpu, interfaces, auth
├── web/ # access_combined
├── retail/ # orders
├── servicebus/ # azure servicebus
├── erp/ # sap
└── itsm/ # servicenow
Splunk applies the FAKE: prefix to all sourcetypes at index time via props.conf/transforms.conf. Use the prefixed form in SPL:
index=fake_tshrt demo_id=exfil | stats count by sourcetype
index=fake_tshrt sourcetype="FAKE:cisco:asa" | timechart count by action
index=fake_tshrt sourcetype="FAKE:perfmon" demo_host="SQL-PROD-01" | timechart avg(Value) by counter
index=fake_tshrt sourcetype="FAKE:cisco:cybervision:events" demo_id=ot_rogue_device
All scenario events carry demo_id=<scenario> for easy filtering.