メニュー
CQL pattern catalog — curated detection engineering patterns for CrowdStrike NG-SIEM. Use when writing, reviewing, or debugging CQL queries.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Analyze CrowdStrike NGSIEM detections for tuning opportunities based on environmental context, recent false positives, and available enrichment functions. Use when tuning detections (including behavioral rules with correlate()), reducing false positives, enhancing detection coverage, or reviewing OOTB templates for production deployment.
Unified SOC analyst workflow for CrowdStrike NGSIEM — triage alerts, investigate security events, hunt threats, and tune detections. Use when triaging alerts, investigating detections, running daily SOC review, or tuning for false positives.
Threat-model-first detection planning for data sources without OOTB coverage. Analyzes what threats apply to a source type, validates against live log data, and produces a prioritized detection backlog with handoff docs for authoring skills. Use when onboarding a new data source, planning detection coverage for a source without OOTB templates, or assessing what threats a source can detect.
Autonomous threat hunting using the PEAK framework (Prepare → Execute → Act). Executes hypothesis-driven, intelligence-driven, and baseline hunts against CrowdStrike NG-SIEM. Produces hunt reports, detection backlogs, and visibility gap reports. Use when proactively hunting for threats, validating detection coverage, or responding to new threat intelligence.
Detection-to-response mapping and SOAR playbook design. Analyzes detections, recommends tiered response actions (observe, investigate, contain, remediate), and produces handoff docs for fusion-workflows to generate workflow YAML. Use when planning response automation for detections, designing SOAR playbooks, or mapping detections to Falcon Fusion workflow actions.
Design multi-event behavioral detection rules using CrowdStrike NG-SIEM correlate() function. Use when building attack chain detections, correlating multiple events across time windows, or creating behavioral rules that detect complex threat patterns across AWS, EntraID, and CrowdStrike data sources.
| name | cql-patterns |
| description | CQL pattern catalog — curated detection engineering patterns for CrowdStrike NG-SIEM. Use when writing, reviewing, or debugging CQL queries. |
| allowed-tools | Read, Grep, Glob |
Curated, battle-tested CQL patterns for detection engineering in CrowdStrike NG-SIEM. This is a pattern catalog, not an API reference — it shows how to combine CQL functions into effective detections. For function-level documentation, see crowdstrike-resources/docs/CQL/.
| Category | File | When to Use |
|---|---|---|
| Correlation | patterns/correlation.md | Multi-event detection: defineTable chains, correlate() sequences, readFile merge |
| Enrichment | patterns/enrichment.md | Adding context: join, selfJoinFilter, match() with CSV lookups, ipLocation, aid_master |
| Aggregation | patterns/aggregation.md | Summarizing events: groupBy with thresholds, bucket() time windows, timeChart, session() |
| String & Decode | patterns/string-and-decode.md | Parsing data: regex named captures, base64Decode, parseXml, bitfield:extractFlags, kvParse |
| Scoring | patterns/scoring.md | Risk assessment: weighted case{} scoring, severity tiering, slidingTimeWindow+rulesHit |
| Baselining | patterns/baselining.md | Anomaly detection: neighbor() sequential analysis, time-window baselines, geography:distance |
| Output | patterns/output.md | Formatting results: table vs select, format() for deep links, unit:convert, formatTime |
Load pattern files based on your current task:
| Task | Load These |
|---|---|
| Writing a new detection | scoring + correlation |
| Hunting / investigation | enrichment + aggregation |
| Tuning an existing detection | baselining + scoring |
| Debugging query output | output + string-and-decode |
| All tasks | Read this entry point first for global gotchas below |
Critical pitfalls that apply across all CQL work. Read these before writing any query.
# prefix REQUIRED for tagged fields. #event_simpleName=DnsRequest not event_simpleName=DnsRequest — unprefixed silently returns 0 results with no error.
Query optimization order. Put cheap filters first, expensive operations last:
time filter -> tag filter -> field filter -> negative filter -> regex -> functions -> aggregation -> rename -> join -> view
table() vs select(). table() creates a new result set and is an aggregation (limits to 200 rows by default). select() picks fields from existing results without limiting rows. Use select() unless you specifically need aggregation behavior.
String comparison is case-sensitive by default. Use /pattern/i for case-insensitive matching.
join() default mode is inner. Use mode=left to preserve all events from the base query when the join table has no match.
IP enrichment chain order. Run in this sequence — each adds fields the next can use:
ipLocation() -> asn() -> rdns(). Run these after groupBy() so they execute once per unique IP, not once per raw event.
Saved search description limit: 2000 characters. Keep description brief; put full documentation in queryString comments.
NG-SIEM query timeout: ~120 seconds. Break complex queries into stages using defineTable to avoid timeouts.
Pre-calculate arithmetic. Do division/multiplication before case/test blocks, not inside them.
Handle null baselines. Always handle missing data with case { field!=* | default; * } or default().
Profile with explain:asTable(). Append to any query to get per-stage performance metrics (timeMs, event counts, prefilter effectiveness). Use it to validate optimization order and find bottlenecks. Ad hoc only — do not include in scheduled searches, triggers, or dashboards. Not supported with correlate(). See output patterns for details.
For complete CQL function API documentation, see crowdstrike-resources/docs/CQL/docs/CrowdStrike-Query-Language/combined.md or individual function files in that directory.
