Install this skill
npx skills add mukul975/Anthropic-Cybersecurity-SkillsFramework mappings
MITRE ATT&CK
T1046 on the official MITRE ATT&CK siteT1053 on the official MITRE ATT&CK siteT1053.003 on the official MITRE ATT&CK siteT1053.005 on the official MITRE ATT&CK siteT1057 on the official MITRE ATT&CK siteT1078 on the official MITRE ATT&CK siteT1082 on the official MITRE ATT&CK siteT1083 on the official MITRE ATT&CK siteT1547 on the official MITRE ATT&CK site
When to Use
- When proactively hunting for persistence mechanisms in Windows environments
- After detecting schtasks.exe or at.exe usage in process creation logs
- When investigating malware that survives reboots and user logoffs
- During incident response to enumerate all persistence on compromised systems
- When Windows Security Event ID 4698 (Scheduled Task Created) fires for unusual tasks
Prerequisites
- Windows Security Event ID 4698/4699/4702 (Task Created/Deleted/Updated)
- Sysmon Event ID 1 for schtasks.exe process creation with command lines
- Windows Task Scheduler operational log (Microsoft-Windows-TaskScheduler/Operational)
- PowerShell logging for Register-ScheduledTask cmdlet usage
- Access to Task Scheduler XML definitions on endpoints
Workflow
- Enumerate All Scheduled Tasks: Collect complete task inventory from target systems using
schtasks /query /fo CSV /vorGet-ScheduledTaskPowerShell cmdlet. - Monitor Task Creation Events: Track Event ID 4698 for new task creation, correlating with the creating process and user account context.
- Analyze Task Actions: Examine what each task executes. Flag tasks running scripts (PowerShell, cmd, wscript), binaries from user-writable paths (TEMP, AppData, Downloads), or encoded/obfuscated commands.
- Check Task Triggers: Review trigger conditions. Tasks triggered by system startup, user logon, or short intervals (1-5 minutes) warrant investigation.
- Identify Hidden or Disguised Tasks: Hunt for tasks with names mimicking legitimate Windows tasks, tasks with Security Descriptor modifications hiding them from standard enumeration, or tasks stored in non-standard registry locations.
- Correlate with Process Execution: Match scheduled task execution events with process creation logs to confirm what actually runs.
- Baseline and Diff: Compare current task inventory against known-good baselines to identify new, modified, or unexpected tasks.
Detection Queries
Splunk -- Scheduled Task Creation
index=wineventlog EventCode=4698
| spath output=TaskName path=EventData.TaskName
| spath output=TaskContent path=EventData.TaskContent
| where NOT match(TaskName, "(?i)(\\\\Microsoft\\\\|\\\\Windows\\\\)")
| table _time Computer SubjectUserName TaskName TaskContentSplunk -- Schtasks.exe Suspicious Usage
index=sysmon EventCode=1 Image="*\\schtasks.exe"
| where match(CommandLine, "(?i)/create")
| where match(CommandLine, "(?i)(powershell|cmd|wscript|cscript|mshta|rundll32|regsvr32|http|https|\\\\temp\\\\|\\\\appdata\\\\)")
| table _time Computer User CommandLine ParentImageKQL -- Microsoft Sentinel
SecurityEvent
| where EventID == 4698
| extend TaskName = tostring(EventData.TaskName)
| extend TaskContent = tostring(EventData.TaskContent)
| where TaskContent has_any ("powershell", "cmd.exe", "wscript", "http://", "https://", "\\Temp\\", "\\AppData\\")
| project TimeGenerated, Computer, Account, TaskName, TaskContentCommon Scenarios
- Cobalt Strike Persistence: Creates scheduled tasks via schtasks.exe to execute PowerShell download cradles at user logon intervals.
- Ransomware Staging: Task created to run encryption payload at a future time, often during off-hours for maximum impact.
- Hidden Task via SD Modification: Attacker modifies Security Descriptor of scheduled task to hide it from normal enumeration while maintaining execution.
- COM Handler Abuse: Task uses COM handler rather than direct executable path, making action inspection more complex.
- Lateral Movement via Tasks: Remote scheduled task creation using
schtasks /create /s REMOTE_HOSTfor execution on other systems.
Output Format
Hunt ID: TH-SCHTASK-[DATE]-[SEQ]
Host: [Hostname]
Task Name: [Full task path]
Action: [Command/Script executed]
Trigger: [Startup/Logon/Timer/Event]
Created By: [User account]
Created From: [Local/Remote]
Creation Time: [Timestamp]
Run As: [Execution account]
Risk Level: [Critical/High/Medium/Low]Source materials
References and resources
Everything below is rendered for inspection. Script files are read-only and never run.
References 3
api-reference.md1.9 KB
API Reference: Hunting for Suspicious Scheduled Tasks
Windows Event IDs
| Event ID | Source | Description |
|---|---|---|
| 4698 | Security | Scheduled task created |
| 4699 | Security | Scheduled task deleted |
| 4702 | Security | Scheduled task updated |
| 106 | TaskScheduler | Task registered |
| 200/201 | TaskScheduler | Task executed / completed |
python-evtx
import Evtx.Evtx as evtx
import xml.etree.ElementTree as ET
with evtx.Evtx("Security.evtx") as log:
for record in log.records():
root = ET.fromstring(record.xml())
ns = {"ns": "http://schemas.microsoft.com/win/2004/08/events/event"}
eid = root.find(".//ns:EventID", ns).text
if eid == "4698":
data = {d.get("Name"): d.text
for d in root.findall(".//ns:Data", ns)}Splunk SPL
index=wineventlog EventCode=4698
| spath output=TaskName path=EventData.TaskName
| spath output=TaskContent path=EventData.TaskContent
| where NOT match(TaskName, "\\\\Microsoft\\\\Windows\\\\")
| where match(TaskContent, "(?i)(powershell|cmd|wscript|http)")
| table _time Computer SubjectUserName TaskName TaskContentKQL (Microsoft Sentinel)
SecurityEvent
| where EventID == 4698
| extend TaskContent = tostring(EventData.TaskContent)
| where TaskContent has_any ("powershell", "cmd.exe", "Temp", "AppData")
| project TimeGenerated, Computer, Account, TaskContentPowerShell Enumeration
Get-ScheduledTask | Where-Object {
$_.Actions.Execute -match 'powershell|cmd|wscript' -or
$_.Actions.Execute -match '\\Temp\\|\\AppData\\'
} | Select-Object TaskName, TaskPath, @{N='Action';E={$_.Actions.Execute}}References
- MITRE T1053.005: https://attack.mitre.org/techniques/T1053/005/
- python-evtx: https://github.com/williballenthin/python-evtx
- Sigma rules for schtasks: https://github.com/SigmaHQ/sigma
standards.md1.4 KB
Standards and References - Suspicious Scheduled Tasks
MITRE ATT&CK References
| Technique | Name | Usage |
|---|---|---|
| T1053.005 | Scheduled Task | Primary persistence/execution technique |
| T1053.003 | Cron (Linux) | Scheduled execution on Linux |
| T1078 | Valid Accounts | Tasks running under legitimate accounts |
Windows Event IDs
| Event ID | Source | Description |
|---|---|---|
| 4698 | Security | Scheduled task created |
| 4699 | Security | Scheduled task deleted |
| 4700 | Security | Scheduled task enabled |
| 4701 | Security | Scheduled task disabled |
| 4702 | Security | Scheduled task updated |
| 106 | TaskScheduler/Operational | Task registered |
| 200 | TaskScheduler/Operational | Action started |
| 201 | TaskScheduler/Operational | Action completed |
Suspicious Task Indicators
| Indicator | Description |
|---|---|
| User-writable paths | Actions executing from TEMP, AppData, Downloads |
| Encoded commands | Base64 or -EncodedCommand in arguments |
| Script interpreters | PowerShell, cmd, wscript, cscript as actions |
| Short intervals | Trigger repeating every 1-5 minutes |
| System startup trigger | Task runs at boot for persistence |
| Remote creation | Task created from remote system |
| Name mimicry | Task name similar to legitimate Windows tasks |
| Hidden SD | Security Descriptor modified to hide task |
workflows.md1.2 KB
Detailed Hunting Workflow - Suspicious Scheduled Tasks
Phase 1: Task Enumeration
# Full task export with details
Get-ScheduledTask | Where-Object { $_.TaskPath -notmatch "\\Microsoft\\" } |
ForEach-Object { $_ | Get-ScheduledTaskInfo; $_.Actions | Select-Object Execute, Arguments }
# Check for hidden tasks in registry
Get-ChildItem "HKLM:\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Schedule\TaskCache\Tree" -RecursePhase 2: SIEM Analysis
index=wineventlog EventCode=4698
| spath output=TaskName path=EventData.TaskName
| spath output=TaskContent path=EventData.TaskContent
| rex field=TaskContent "<Command>(?<cmd>[^<]+)</Command>"
| rex field=TaskContent "<Arguments>(?<args>[^<]+)</Arguments>"
| table _time Computer SubjectUserName TaskName cmd argsPhase 3: Remote Task Creation Detection
index=sysmon EventCode=1 Image="*\\schtasks.exe"
| where match(CommandLine, "(?i)/create.*/s\s+")
| rex field=CommandLine "/s\s+(?<remote_host>\S+)"
| table _time Computer User remote_host CommandLinePhase 4: Response
- Remove malicious scheduled tasks
- Check task XML definitions for hidden parameters
- Audit all non-Microsoft scheduled tasks across fleet
- Deploy detection rules for suspicious task creation
Scripts 2
agent.py6.8 KB
Display-only source. This catalog never executes bundled scripts.
#!/usr/bin/env python3
"""Agent for hunting suspicious scheduled tasks on Windows endpoints."""
import json
import argparse
import re
import xml.etree.ElementTree as ET
from datetime import datetime
try:
import Evtx.Evtx as evtx
except ImportError:
evtx = None
SUSPICIOUS_ACTIONS = [
r"powershell", r"pwsh", r"cmd\.exe.*/c", r"wscript", r"cscript",
r"mshta", r"rundll32", r"regsvr32", r"certutil",
r"bitsadmin", r"msiexec.*http",
]
SUSPICIOUS_PATHS = [
r"\\temp\\", r"\\tmp\\", r"\\appdata\\", r"\\downloads\\",
r"\\public\\", r"\\programdata\\", r"\\users\\.*\\desktop\\",
r"c:\\windows\\temp",
]
LEGITIMATE_TASK_PREFIXES = [
"\\Microsoft\\Windows\\", "\\Microsoft\\Office\\",
"\\Microsoft\\EdgeUpdate\\",
]
def parse_schtasks_csv(csv_path):
"""Parse output of schtasks /query /fo CSV /v."""
import csv
tasks = []
with open(csv_path, newline="", encoding="utf-8-sig") as f:
reader = csv.DictReader(f)
for row in reader:
tasks.append({
"hostname": row.get("HostName", ""),
"task_name": row.get("TaskName", ""),
"status": row.get("Status", ""),
"task_to_run": row.get("Task To Run", ""),
"run_as_user": row.get("Run As User", ""),
"schedule_type": row.get("Schedule Type", ""),
"author": row.get("Author", ""),
})
return tasks
def analyze_tasks(tasks):
"""Analyze scheduled tasks for suspicious properties."""
findings = []
for task in tasks:
name = task.get("task_name", "")
action = task.get("task_to_run", "")
run_as = task.get("run_as_user", "")
is_legit_prefix = any(name.startswith(p) for p in LEGITIMATE_TASK_PREFIXES)
risk_score = 0
reasons = []
for pattern in SUSPICIOUS_ACTIONS:
if re.search(pattern, action, re.IGNORECASE):
risk_score += 30
reasons.append(f"suspicious_action:{pattern}")
for pattern in SUSPICIOUS_PATHS:
if re.search(pattern, action, re.IGNORECASE):
risk_score += 25
reasons.append(f"suspicious_path:{pattern}")
if run_as and "SYSTEM" in run_as.upper():
risk_score += 15
reasons.append("runs_as_system")
if not is_legit_prefix and name.count("\\") <= 1:
risk_score += 10
reasons.append("non_standard_location")
if re.search(r"(http|https|ftp)://", action, re.IGNORECASE):
risk_score += 40
reasons.append("network_url_in_action")
if re.search(r"-enc\s|encodedcommand", action, re.IGNORECASE):
risk_score += 35
reasons.append("encoded_command")
if risk_score > 0:
severity = "CRITICAL" if risk_score >= 60 else "HIGH" if risk_score >= 30 else "MEDIUM"
findings.append({
"task_name": name,
"action": action[:500],
"run_as_user": run_as,
"risk_score": risk_score,
"severity": severity,
"reasons": reasons,
"hostname": task.get("hostname", ""),
})
return sorted(findings, key=lambda x: x["risk_score"], reverse=True)
def hunt_evtx_4698(evtx_path):
"""Hunt Event ID 4698 (scheduled task creation) in EVTX."""
if evtx is None:
return []
findings = []
with evtx.Evtx(evtx_path) as log:
for record in log.records():
xml_str = record.xml()
try:
root = ET.fromstring(xml_str)
ns = {"ns": "http://schemas.microsoft.com/win/2004/08/events/event"}
eid_el = root.find(".//ns:EventID", ns)
if eid_el is None or eid_el.text != "4698":
continue
data = {}
for d in root.findall(".//ns:Data", ns):
data[d.get("Name", "")] = d.text or ""
task_name = data.get("TaskName", "")
task_content = data.get("TaskContent", "")
is_suspicious = any(
re.search(p, task_content, re.IGNORECASE) for p in SUSPICIOUS_ACTIONS)
if is_suspicious:
findings.append({
"timestamp": record.timestamp().isoformat(),
"task_name": task_name,
"user": data.get("SubjectUserName", ""),
"task_content_preview": task_content[:500],
"severity": "HIGH",
})
except ET.ParseError:
continue
return findings
def generate_sigma_rule():
"""Generate Sigma rule for suspicious scheduled task creation."""
return {
"title": "Suspicious Scheduled Task Created",
"id": "e4db2c6a-3f1b-4c8d-9e2a-7b5c4d6e8f0a",
"status": "production",
"level": "high",
"logsource": {"product": "windows", "service": "security"},
"detection": {
"selection": {"EventID": 4698},
"filter_legit": {"TaskName|startswith": ["\\Microsoft\\Windows\\", "\\Microsoft\\Office\\"]},
"suspicious_content": {
"TaskContent|contains": ["powershell", "cmd /c", "wscript", "mshta",
"\\Temp\\", "\\AppData\\", "http://", "https://"],
},
"condition": "selection and not filter_legit and suspicious_content",
},
"tags": ["attack.persistence", "attack.execution", "attack.t1053.005"],
}
def main():
parser = argparse.ArgumentParser(description="Suspicious Scheduled Tasks Hunter")
parser.add_argument("--csv", help="schtasks CSV export")
parser.add_argument("--evtx", help="Security EVTX log file")
parser.add_argument("--output", default="schtask_hunt_report.json")
parser.add_argument("--action", choices=["analyze", "hunt_evtx", "sigma", "full"],
default="full")
args = parser.parse_args()
report = {"generated_at": datetime.utcnow().isoformat(), "findings": {}}
if args.action in ("analyze", "full") and args.csv:
tasks = parse_schtasks_csv(args.csv)
findings = analyze_tasks(tasks)
report["findings"]["task_analysis"] = findings
print(f"[+] Suspicious tasks: {len(findings)} / {len(tasks)} total")
if args.action in ("hunt_evtx", "full") and args.evtx:
findings = hunt_evtx_4698(args.evtx)
report["findings"]["evtx_4698"] = findings
print(f"[+] EVTX 4698 suspicious: {len(findings)}")
if args.action in ("sigma", "full"):
report["findings"]["sigma_rule"] = generate_sigma_rule()
print("[+] Sigma rule generated")
with open(args.output, "w") as f:
json.dump(report, f, indent=2, default=str)
print(f"[+] Report saved to {args.output}")
if __name__ == "__main__":
main()
process.py4.7 KB
Display-only source. This catalog never executes bundled scripts.
#!/usr/bin/env python3
"""
Suspicious Scheduled Task Detection Script
Analyzes Windows event logs for malicious scheduled task creation,
modification, and execution patterns.
"""
import json
import csv
import argparse
import datetime
import re
from pathlib import Path
SUSPICIOUS_ACTION_PATTERNS = [
(r"(?i)powershell", "powershell_execution", "HIGH"),
(r"(?i)(cmd\.exe|cmd\s/c)", "command_shell", "HIGH"),
(r"(?i)(wscript|cscript)", "script_execution", "HIGH"),
(r"(?i)(mshta|rundll32|regsvr32)", "lolbin_execution", "CRITICAL"),
(r"(?i)(http|https)://", "network_download", "CRITICAL"),
(r"(?i)(\\temp\\|\\appdata\\|\\downloads\\)", "user_writable_path", "HIGH"),
(r"(?i)(-enc\s|-encodedcommand)", "encoded_command", "CRITICAL"),
(r"(?i)(base64|frombase64)", "base64_content", "HIGH"),
]
LEGITIMATE_TASK_PATHS = [
r"(?i)\\Microsoft\\",
r"(?i)\\Windows\\",
r"(?i)\\Adobe\\",
r"(?i)\\Google\\",
]
def parse_events(input_path: str) -> list[dict]:
path = Path(input_path)
events = []
if path.suffix == ".json":
with open(path, "r", encoding="utf-8") as f:
data = json.load(f)
events = data if isinstance(data, list) else data.get("events", [])
elif path.suffix == ".csv":
with open(path, "r", encoding="utf-8-sig") as f:
events = [dict(row) for row in csv.DictReader(f)]
return events
def detect_suspicious_tasks(events: list[dict]) -> list[dict]:
findings = []
for event in events:
event_code = str(event.get("EventCode", event.get("EventID", "")))
computer = event.get("Computer", event.get("host", ""))
timestamp = event.get("TimeCreated", event.get("_time", ""))
user = event.get("SubjectUserName", event.get("user", ""))
if event_code == "4698":
task_name = event.get("TaskName", "")
task_content = event.get("TaskContent", "")
if any(re.search(p, task_name) for p in LEGITIMATE_TASK_PATHS):
continue
cmd_match = re.search(r"<Command>([^<]+)</Command>", task_content)
args_match = re.search(r"<Arguments>([^<]+)</Arguments>", task_content)
command = cmd_match.group(1) if cmd_match else ""
arguments = args_match.group(1) if args_match else ""
full_action = f"{command} {arguments}"
for pattern, category, severity in SUSPICIOUS_ACTION_PATTERNS:
if re.search(pattern, full_action):
findings.append({
"timestamp": timestamp, "computer": computer, "user": user,
"task_name": task_name, "command": command,
"arguments": arguments, "category": category,
"severity": severity, "technique": "T1053.005",
})
break
elif event_code == "1":
image = event.get("Image", "")
cmdline = event.get("CommandLine", "")
if "schtasks.exe" in image.lower() and "/create" in cmdline.lower():
for pattern, category, severity in SUSPICIOUS_ACTION_PATTERNS:
if re.search(pattern, cmdline):
findings.append({
"timestamp": timestamp, "computer": computer, "user": user,
"task_name": "via_schtasks",
"command": image, "arguments": cmdline,
"category": f"schtasks_{category}",
"severity": severity, "technique": "T1053.005",
})
break
return sorted(findings, key=lambda x: {"CRITICAL": 0, "HIGH": 1, "MEDIUM": 2}.get(x["severity"], 3))
def run_hunt(input_path: str, output_dir: str) -> None:
print(f"[*] Scheduled Task Hunt - {datetime.datetime.now().isoformat()}")
events = parse_events(input_path)
findings = detect_suspicious_tasks(events)
print(f"[!] Suspicious task detections: {len(findings)}")
output_path = Path(output_dir)
output_path.mkdir(parents=True, exist_ok=True)
with open(output_path / "schtask_findings.json", "w", encoding="utf-8") as f:
json.dump({"hunt_id": f"TH-SCHTASK-{datetime.date.today().isoformat()}",
"findings": findings}, f, indent=2)
print(f"[+] Results written to {output_dir}")
def main():
parser = argparse.ArgumentParser(description="Suspicious Scheduled Task Detection")
parser.add_argument("--input", "-i", required=True)
parser.add_argument("--output", "-o", default="./schtask_hunt_output")
args = parser.parse_args()
run_hunt(args.input, args.output)
if __name__ == "__main__":
main()
Assets 1
template.mdtext/markdown · 0.8 KBKeep exploring