managing-crowdstrike

name	managing-crowdstrike
description	Use when working with Crowdstrike — crowdStrike Falcon endpoint detection and response, threat intelligence, host management, and incident investigation. Covers detection queries, host inventory, IOC searches, real-time response sessions, and vulnerability assessment. Use when investigating security incidents, reviewing endpoint health, analyzing threat detections, or managing CrowdStrike Falcon sensors.
connection_type	crowdstrike
preload	false

CrowdStrike Falcon Management Skill

Manage and analyze CrowdStrike Falcon detections, hosts, incidents, and threat intelligence.

API Conventions

Authentication

All API calls use OAuth2 bearer tokens via Authorization: Bearer $CROWDSTRIKE_ACCESS_TOKEN -- injected automatically. Never hardcode tokens.

Base URL

https://api.crowdstrike.com

Core Helper Function

#!/bin/bash

cs_api() {
    local method="$1"
    local endpoint="$2"
    local data="${3:-}"

    if [ -n "$data" ]; then
        curl -s -X "$method" \
            -H "Authorization: Bearer $CROWDSTRIKE_ACCESS_TOKEN" \
            -H "Content-Type: application/json" \
            "https://api.crowdstrike.com${endpoint}" \
            -d "$data"
    else
        curl -s -X "$method" \
            -H "Authorization: Bearer $CROWDSTRIKE_ACCESS_TOKEN" \
            -H "Content-Type: application/json" \
            "https://api.crowdstrike.com${endpoint}"
    fi
}

Output Rules

• TOKEN EFFICIENCY: Extract only needed fields with jq
• Target <=50 lines per script output
• Always filter by relevant time range to avoid huge response sets
• Never dump full API responses

Discovery Phase

#!/bin/bash
echo "=== Host Inventory Summary ==="
cs_api GET "/devices/queries/devices-scroll/v1?limit=5000" \
    | jq '.meta.pagination.total as $total | "Total managed hosts: \($total)"' -r

echo ""
echo "=== Recent Detections (last 24h) ==="
cs_api GET "/detects/queries/detects/v1?filter=created_timestamp:>'$(date -u -v-1d +%Y-%m-%dT%H:%M:%SZ)'&limit=100" \
    | jq '.meta.pagination.total as $total | "Detections (24h): \($total)"' -r

echo ""
echo "=== Active Incidents ==="
cs_api GET "/incidents/queries/incidents/v1?filter=status:['20','25']&limit=50" \
    | jq '.meta.pagination.total as $total | "Open incidents: \($total)"' -r

Analysis Phase

Detection Overview

#!/bin/bash
echo "=== Critical/High Detections ==="
IDS=$(cs_api GET "/detects/queries/detects/v1?filter=max_severity_displayname:['Critical','High']+created_timestamp:>'$(date -u -v-7d +%Y-%m-%dT%H:%M:%SZ)'&limit=20" | jq -r '.resources[:20] | join(",")')

if [ -n "$IDS" ]; then
    cs_api POST "/detects/entities/summaries/GET/v1" "{\"ids\":[$(echo "$IDS" | sed 's/,/","/g' | sed 's/^/"/;s/$/"/')]}" \
        | jq -r '.resources[] | "\(.created_timestamp[0:16])\t\(.max_severity_displayname)\t\(.status)\t\(.device.hostname)\t\(.detection_id)"' \
        | column -t | head -20
fi

echo ""
echo "=== Detection Count by Severity ==="
cs_api GET "/detects/aggregates/detects/GET/v1" \
    -d '{"date_ranges":[{"from":"'"$(date -u -v-7d +%Y-%m-%dT%H:%M:%SZ)"'","to":"'"$(date -u +%Y-%m-%dT%H:%M:%SZ)"'"}],"field":"max_severity_displayname","type":"terms"}' \
    | jq -r '.resources[].buckets[] | "\(.label): \(.count)"'

Host Health

#!/bin/bash
echo "=== Hosts by Platform ==="
cs_api POST "/devices/aggregates/devices/GET/v1" \
    '[{"type":"terms","field":"platform_name"}]' \
    | jq -r '.resources[].buckets[] | "\(.label): \(.count)"'

echo ""
echo "=== Hosts with Stale Sensors (>7 days) ==="
cs_api GET "/devices/queries/devices/v1?filter=last_seen:<='$(date -u -v-7d +%Y-%m-%dT%H:%M:%SZ)'&limit=10" \
    | jq '.meta.pagination.total as $total | "Stale sensors: \($total)"' -r

echo ""
echo "=== Sensor Versions ==="
cs_api POST "/devices/aggregates/devices/GET/v1" \
    '[{"type":"terms","field":"agent_version","size":10}]' \
    | jq -r '.resources[].buckets[] | "\(.label): \(.count)"' | head -10

Incident Investigation

#!/bin/bash
INCIDENT_ID="${1:?Incident ID required}"

echo "=== Incident Details ==="
cs_api POST "/incidents/entities/incidents/GET/v1" "{\"ids\":[\"$INCIDENT_ID\"]}" \
    | jq '.resources[0] | {
        incident_id: .incident_id,
        status: .status,
        severity: .fine_score,
        start: .start,
        end: .end,
        hosts: (.hosts | length),
        tactics: .tactics,
        techniques: .techniques
    }'

echo ""
echo "=== Associated Detections ==="
cs_api GET "/incidents/queries/behaviors/v1?filter=incident_id:'$INCIDENT_ID'" \
    | jq -r '.resources[:10][]'

Output Format

Present results as a structured report:

Managing Crowdstrike Report
═══════════════════════════
Resources discovered: [count]

Resource       Status    Key Metric    Issues
──────────────────────────────────────────────
[name]         [ok/warn] [value]       [findings]

Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]

Target ≤50 lines of output. Use tables for multi-resource comparisons.

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.

Counter-Rationalizations

Shortcut	Counter	Why
"I'll skip discovery and check known resources"	Always run Phase 1 discovery first	Resource names change, new resources appear — assumed names cause errors
"The user only asked for a quick check"	Follow the full discovery → analysis flow	Quick checks miss critical issues; structured analysis catches silent failures
"Default configuration is probably fine"	Audit configuration explicitly	Defaults often leave logging, security, and optimization features disabled
"Metrics aren't needed for this"	Always check relevant metrics when available	API/CLI responses show current state; metrics reveal trends and intermittent issues
"I don't have access to that"	Try the command and report the actual error	Assumed permission failures prevent useful investigation; actual errors are informative

Common Pitfalls

• Two-step entity lookup: Query endpoints return IDs only -- must call entity endpoints to get details
• FQL syntax: Filter queries use Falcon Query Language with single quotes for values
• Rate limits: 6000 requests/min for most endpoints -- stagger parallel calls
• Pagination: Default limit varies by endpoint -- always specify limit parameter
• OAuth2 token refresh: Tokens expire after 30 minutes -- refresh via /oauth2/token