npx skills add mukul975/Anthropic-Cybersecurity-SkillsMITRE ATT&CK
When to Use
- When determining which programs were executed on a Windows system and when
- During malware investigations to confirm execution of suspicious binaries
- For establishing a timeline of application usage during an incident
- When correlating program execution with other forensic artifacts
- To identify anti-forensic tools or unauthorized software that was run
Prerequisites
- Access to Windows Prefetch directory (C:\Windows\Prefetch) from forensic image
- PECmd (Eric Zimmerman), WinPrefetchView, or python-prefetch parser
- Understanding of Prefetch file format (versions 17, 23, 26, 30)
- Windows system with Prefetch enabled (default on client OS, disabled on servers)
- Knowledge of Prefetch naming conventions (APPNAME-HASH.pf)
Workflow
Step 1: Extract Prefetch Files from Forensic Image
# Mount the forensic image
mount -o ro,loop,offset=$((2048*512)) /cases/case-2024-001/images/evidence.dd /mnt/evidence
# Copy all prefetch files
mkdir -p /cases/case-2024-001/prefetch/
cp /mnt/evidence/Windows/Prefetch/*.pf /cases/case-2024-001/prefetch/
# Count and list prefetch files
ls -la /cases/case-2024-001/prefetch/ | wc -l
ls -la /cases/case-2024-001/prefetch/ | head -30
# Hash all prefetch files for integrity
sha256sum /cases/case-2024-001/prefetch/*.pf > /cases/case-2024-001/prefetch/pf_hashes.txt
# Note: Prefetch filename format is EXECUTABLE_NAME-XXXXXXXX.pf
# The hash (XXXXXXXX) is based on the executable path
# Same executable from different paths creates different prefetch filesStep 2: Parse Prefetch Files with PECmd
# Using Eric Zimmerman's PECmd (Windows or via Mono/Wine on Linux)
# Download from https://ericzimmerman.github.io/
# Parse a single prefetch file
PECmd.exe -f "C:\cases\prefetch\POWERSHELL.EXE-A]B2C3D4.pf"
# Parse all prefetch files and output to CSV
PECmd.exe -d "C:\cases\prefetch\" --csv "C:\cases\analysis\" --csvf prefetch_results.csv
# Parse with JSON output
PECmd.exe -d "C:\cases\prefetch\" --json "C:\cases\analysis\" --jsonf prefetch_results.json
# Output includes for each file:
# - Executable name and path
# - Run count
# - Last run time (up to 8 timestamps in Windows 10)
# - Files and directories referenced during execution
# - Volume information (serial number, creation date)
# - Prefetch file creation timeStep 3: Parse with Python for Linux-Based Analysis
pip install prefetch
python3 << 'PYEOF'
import os
import json
from datetime import datetime
# Parse prefetch files using python
import struct
def parse_prefetch(filepath):
"""Parse a Windows Prefetch file."""
with open(filepath, 'rb') as f:
data = f.read()
# Check for MAM compressed format (Windows 10)
if data[:4] == b'MAM\x04':
import lznt1 # or use DecompressBuffer
# Windows 10 prefetch files are compressed
print(f" [Compressed Win10 format - use PECmd for full parsing]")
return None
# Version 17 (XP), 23 (Vista/7), 26 (8.1), 30 (10)
version = struct.unpack('<I', data[0:4])[0]
signature = data[4:8]
if signature != b'SCCA':
print(f" Invalid prefetch signature")
return None
file_size = struct.unpack('<I', data[8:12])[0]
exec_name = data[16:76].decode('utf-16-le').strip('\x00')
run_count = struct.unpack('<I', data[208:212])[0] if version >= 23 else struct.unpack('<I', data[144:148])[0]
result = {
'version': version,
'executable': exec_name,
'file_size': file_size,
'run_count': run_count,
}
# Extract last execution timestamps
if version == 23: # Vista/7 - 1 timestamp
ts = struct.unpack('<Q', data[128:136])[0]
result['last_run'] = filetime_to_datetime(ts)
elif version >= 26: # Win8+ - up to 8 timestamps
timestamps = []
for i in range(8):
ts = struct.unpack('<Q', data[128+i*8:136+i*8])[0]
if ts > 0:
timestamps.append(filetime_to_datetime(ts))
result['last_run_times'] = timestamps
return result
def filetime_to_datetime(ft):
"""Convert Windows FILETIME to datetime string."""
if ft == 0:
return None
timestamp = (ft - 116444736000000000) / 10000000
try:
return datetime.utcfromtimestamp(timestamp).strftime('%Y-%m-%d %H:%M:%S UTC')
except (OSError, ValueError):
return None
# Process all prefetch files
prefetch_dir = '/cases/case-2024-001/prefetch/'
results = []
for filename in sorted(os.listdir(prefetch_dir)):
if filename.lower().endswith('.pf'):
filepath = os.path.join(prefetch_dir, filename)
print(f"\n=== {filename} ===")
result = parse_prefetch(filepath)
if result:
print(f" Executable: {result['executable']}")
print(f" Run Count: {result['run_count']}")
if 'last_run' in result:
print(f" Last Run: {result['last_run']}")
elif 'last_run_times' in result:
for i, ts in enumerate(result['last_run_times']):
print(f" Run Time {i+1}: {ts}")
results.append(result)
# Save results
with open('/cases/case-2024-001/analysis/prefetch_analysis.json', 'w') as f:
json.dump(results, f, indent=2)
PYEOFStep 4: Identify Suspicious Execution Evidence
# Search for known malicious tool names in prefetch
ls /cases/case-2024-001/prefetch/ | grep -iE \
'(MIMIKATZ|PSEXEC|WMIC|COBALT|BEACON|PWDUMP|PROCDUMP|LAZAGNE|RUBEUS|BLOODHOUND|SHARPHOUND|CERTUTIL|BITSADMIN)'
# Search for script interpreters (potential malicious execution)
ls /cases/case-2024-001/prefetch/ | grep -iE \
'(POWERSHELL|CMD\.EXE|WSCRIPT|CSCRIPT|MSHTA|REGSVR32|RUNDLL32|MSIEXEC)'
# Search for remote access tools
ls /cases/case-2024-001/prefetch/ | grep -iE \
'(TEAMVIEWER|ANYDESK|LOGMEIN|VNC|SPLASHTOP|SCREENCONNECT|AMMYY)'
# Search for data exfiltration tools
ls /cases/case-2024-001/prefetch/ | grep -iE \
'(RAR|7Z|ZIP|RCLONE|MEGA|DROPBOX|ONEDRIVE|GDRIVE|FTP|CURL|WGET)'
# Find recently created prefetch files (newest executables run)
ls -lt /cases/case-2024-001/prefetch/ | head -20
# Cross-reference with Shimcache and Amcache for confirmation
# Prefetch existence = program was executed at least onceStep 5: Build Execution Timeline
# Create timeline from prefetch data
python3 << 'PYEOF'
import json
import csv
with open('/cases/case-2024-001/analysis/prefetch_analysis.json') as f:
data = json.load(f)
timeline = []
for entry in data:
if 'last_run_times' in entry:
for ts in entry['last_run_times']:
if ts:
timeline.append({
'timestamp': ts,
'executable': entry['executable'],
'run_count': entry['run_count'],
'source': 'Prefetch'
})
elif 'last_run' in entry and entry['last_run']:
timeline.append({
'timestamp': entry['last_run'],
'executable': entry['executable'],
'run_count': entry['run_count'],
'source': 'Prefetch'
})
# Sort chronologically
timeline.sort(key=lambda x: x['timestamp'])
# Write timeline CSV
with open('/cases/case-2024-001/analysis/execution_timeline.csv', 'w', newline='') as f:
writer = csv.DictWriter(f, fieldnames=['timestamp', 'executable', 'run_count', 'source'])
writer.writeheader()
writer.writerows(timeline)
# Print suspicious time window
for entry in timeline:
if '2024-01-15' in entry['timestamp'] or '2024-01-16' in entry['timestamp']:
print(f" {entry['timestamp']} | {entry['executable']} (x{entry['run_count']})")
PYEOFKey Concepts
| Concept | Description |
|---|---|
| Prefetch | Windows performance optimization that pre-loads application data and tracks execution |
| SCCA signature | Magic bytes identifying a valid Prefetch file |
| Path hash | CRC-based hash of the executable path forming part of the .pf filename |
| Run count | Number of times the executable has been launched (may wrap around) |
| Last run timestamps | Windows 8+ stores up to 8 most recent execution timestamps |
| Referenced files | List of files and directories accessed during the first 10 seconds of execution |
| Volume information | Drive serial number and creation date identifying the source volume |
| MAM compression | Windows 10 Prefetch files use MAM4 compression requiring decompression before parsing |
Tools & Systems
| Tool | Purpose |
|---|---|
| PECmd | Eric Zimmerman's Prefetch parser with CSV/JSON output |
| WinPrefetchView | NirSoft GUI tool for viewing Prefetch files |
| python-prefetch | Python library for parsing Prefetch files |
| Prefetch Hash Calculator | Tool to calculate expected hash from executable paths |
| KAPE | Automated artifact collection including Prefetch |
| Autopsy | Forensic platform with Prefetch analysis module |
| Plaso/log2timeline | Super-timeline tool that includes Prefetch parser |
| Velociraptor | Endpoint agent with Prefetch collection and analysis artifacts |
Common Scenarios
Scenario 1: Confirming Malware Execution Search Prefetch directory for the malware executable name, confirm execution via Prefetch existence, extract run count and last run time, identify referenced DLLs to understand malware behavior, correlate with registry autorun entries.
Scenario 2: Attacker Tool Usage Timeline Identify Prefetch files for PsExec, Mimikatz, BloodHound, and other attacker tools, build chronological timeline of tool execution, determine the sequence of the attack (reconnaissance, credential theft, lateral movement), match timestamps with network connection logs.
Scenario 3: Data Staging and Exfiltration Look for Prefetch entries of compression tools (7z, WinRAR, zip), identify execution of file transfer utilities (rclone, FTP clients), check for cloud storage client execution, timeline when data staging and transfer occurred.
Scenario 4: Anti-Forensics Detection Check for execution of known anti-forensic tools (CCleaner, Eraser, SDelete), identify if Prefetch directory was recently cleared (fewer files than expected for active system), note timestamps of anti-forensic tool execution relative to other evidence.
Output Format
Prefetch Analysis Summary:
System: Windows 10 Pro (Build 19041)
Prefetch Files: 234
Analysis Period: All available execution history
Execution Statistics:
Total unique executables: 234
First execution: 2023-06-15 (system install)
Latest execution: 2024-01-18 23:45 UTC
Suspicious Executions:
MIMIKATZ.EXE-5F2A3B1C.pf
Run Count: 3 | Last: 2024-01-16 02:30:15 UTC
PSEXEC.EXE-AD70946C.pf
Run Count: 7 | Last: 2024-01-16 02:45:30 UTC
RCLONE.EXE-1F3E5A2B.pf
Run Count: 2 | Last: 2024-01-17 03:15:00 UTC
POWERSHELL.EXE-022A1004.pf
Run Count: 145 | Last: 2024-01-18 14:00:00 UTC
Attack Timeline (from Prefetch):
2024-01-15 14:32 - POWERSHELL.EXE (initial access)
2024-01-16 02:30 - MIMIKATZ.EXE (credential theft)
2024-01-16 02:45 - PSEXEC.EXE (lateral movement)
2024-01-17 03:15 - RCLONE.EXE (data exfiltration)
Report: /cases/case-2024-001/analysis/execution_timeline.csvReferences and resources
Everything below is rendered for inspection. Script files are read-only and never run.
References 1
api-reference.md3.1 KB
API Reference: Windows Prefetch Analysis Tools
Prefetch File Format
Location
C:\Windows\Prefetch\Filename Convention
EXECUTABLE_NAME-XXXXXXXX.pfEXECUTABLE_NAME- Uppercase name of the executed programXXXXXXXX- Hash of the executable path (8 hex characters).pf- Prefetch file extension
Version History
| Version | Windows OS | Notes |
|---|---|---|
| 17 | XP | Basic format |
| 23 | Vista, 7 | Added run count, timestamps |
| 26 | 8, 8.1 | Extended timestamps (8 entries) |
| 30 | 10, 11 | MAM compressed, 8 timestamps |
Header Structure (Uncompressed)
| Offset | Size | Field |
|---|---|---|
| 0 | 4 | Version |
| 4 | 4 | Signature (SCCA) |
| 12 | 4 | File size |
| 16 | 60 | Executable name (UTF-16LE) |
| 76 | 4 | Prefetch hash |
PECmd (Eric Zimmerman) - Full Parser
Syntax
PECmd.exe -f <prefetch_file> # Single file
PECmd.exe -d <prefetch_directory> # Entire directory
PECmd.exe -d <dir> --csv <output_dir> # Export to CSV
PECmd.exe -d <dir> --json <output_dir> # Export to JSON
PECmd.exe -f <file> -q # Quiet modeOutput Fields
| Field | Description |
|---|---|
SourceFilename |
Original executable path |
RunCount |
Number of times executed |
LastRun |
Most recent execution timestamp |
PreviousRun0-7 |
Up to 8 previous run timestamps (Win8+) |
FilesLoaded |
DLLs and files accessed during execution |
Directories |
Directories accessed |
VolumeSerialNumber |
Volume where executable resided |
WinPrefetchView (NirSoft)
GUI Features
- Lists all prefetch files with execution details
- Shows run count, timestamps, referenced files
- Export to CSV, HTML, or text
- Sort by any column for analysis
Python Prefetch Parsing
Structure Parsing
import struct
with open("APP.EXE-HASH.pf", "rb") as f:
data = f.read()
version = struct.unpack_from("<I", data, 0)[0]
signature = data[4:8] # Should be b"SCCA"
exe_name = data[16:76].decode("utf-16-le").rstrip("\x00")
pf_hash = struct.unpack_from("<I", data, 76)[0]FILETIME Conversion
import datetime
def filetime_to_datetime(filetime):
epoch = datetime.datetime(1601, 1, 1)
delta = datetime.timedelta(microseconds=filetime // 10)
return epoch + deltaSuspicious Prefetch Indicators
Offensive Tools
| Tool | Prefetch Name |
|---|---|
| Mimikatz | MIMIKATZ.EXE-*.pf |
| PsExec | PSEXEC.EXE-*.pf, PSEXESVC.EXE-*.pf |
| BloodHound | SHARPHOUND.EXE-*.pf |
| Rubeus | RUBEUS.EXE-*.pf |
| LaZagne | LAZAGNE.EXE-*.pf |
LOLBins (Living Off the Land)
| Binary | Concern |
|---|---|
CERTUTIL.EXE |
File download, Base64 decode |
MSHTA.EXE |
Script execution via HTA |
REGSVR32.EXE |
COM scriptlet execution |
BITSADMIN.EXE |
File download |
MSBUILD.EXE |
Code execution via project files |
Timeline Integration
Plaso / log2timeline
log2timeline.py timeline.plaso /path/to/prefetch/
psort.py -o l2tcsv timeline.plaso > prefetch_timeline.csvScripts 1
agent.py8.0 KB
#!/usr/bin/env python3
"""Windows Prefetch file analysis agent for program execution history forensics."""
import struct
import os
import sys
import datetime
import json
import glob
def parse_prefetch_header(filepath):
"""Parse the Prefetch file header to extract execution metadata."""
with open(filepath, "rb") as f:
data = f.read()
# Check for compression (Windows 10 prefetch files are MAM compressed)
if data[:4] == b"MAM\x04":
# Windows 10 compressed format - need decompression
return {"error": "Compressed prefetch (Windows 10 MAM format) - use PECmd for full parsing",
"compressed": True, "raw_size": len(data)}
# Standard prefetch header (versions 17, 23, 26, 30)
if len(data) < 84:
return {"error": "File too small to be a valid prefetch file"}
version = struct.unpack_from("<I", data, 0)[0]
signature = data[4:8]
if signature != b"SCCA":
return {"error": f"Invalid signature: {signature.hex()} (expected 53434341)"}
file_size = struct.unpack_from("<I", data, 12)[0]
exe_name = data[16:76].decode("utf-16-le", errors="replace").rstrip("\x00")
hash_value = struct.unpack_from("<I", data, 76)[0]
result = {
"version": version,
"signature": signature.hex(),
"file_size": file_size,
"executable_name": exe_name,
"prefetch_hash": f"0x{hash_value:08X}",
}
# Version-specific parsing
if version == 17: # Windows XP
result["format"] = "Windows XP"
run_count = struct.unpack_from("<I", data, 144)[0]
last_run = parse_filetime(struct.unpack_from("<Q", data, 120)[0])
result["run_count"] = run_count
result["last_run_time"] = last_run
elif version == 23: # Windows Vista/7
result["format"] = "Windows Vista/7"
run_count = struct.unpack_from("<I", data, 152)[0]
last_run = parse_filetime(struct.unpack_from("<Q", data, 128)[0])
result["run_count"] = run_count
result["last_run_time"] = last_run
elif version == 26: # Windows 8/8.1
result["format"] = "Windows 8/8.1"
run_count = struct.unpack_from("<I", data, 208)[0]
last_run = parse_filetime(struct.unpack_from("<Q", data, 128)[0])
result["run_count"] = run_count
result["last_run_time"] = last_run
elif version == 30: # Windows 10/11
result["format"] = "Windows 10/11"
result["note"] = "Use PECmd.exe for full Windows 10 prefetch parsing"
else:
result["format"] = f"Unknown version {version}"
return result
def parse_filetime(filetime):
"""Convert Windows FILETIME (100ns intervals since 1601-01-01) to ISO string."""
if filetime == 0:
return "N/A"
try:
epoch = datetime.datetime(1601, 1, 1)
delta = datetime.timedelta(microseconds=filetime // 10)
dt = epoch + delta
return dt.isoformat() + "Z"
except (OverflowError, OSError):
return "Invalid timestamp"
def parse_prefetch_filename(filename):
"""Parse executable name and hash from prefetch filename format: APPNAME-HASH.pf."""
basename = os.path.basename(filename)
if not basename.upper().endswith(".PF"):
return None, None
name_part = basename[:-3] # Remove .pf
parts = name_part.rsplit("-", 1)
if len(parts) == 2:
return parts[0], parts[1]
return name_part, None
def scan_prefetch_directory(prefetch_dir):
"""Scan a directory of prefetch files and extract execution history."""
results = []
pf_files = glob.glob(os.path.join(prefetch_dir, "*.pf"))
pf_files.extend(glob.glob(os.path.join(prefetch_dir, "*.PF")))
for pf_file in sorted(set(pf_files)):
exe_name, pf_hash = parse_prefetch_filename(pf_file)
header = parse_prefetch_header(pf_file)
results.append({
"file": os.path.basename(pf_file),
"parsed_name": exe_name,
"parsed_hash": pf_hash,
"file_modified": datetime.datetime.fromtimestamp(
os.path.getmtime(pf_file)).isoformat(),
"header": header,
})
return results
SUSPICIOUS_EXECUTABLES = [
"MIMIKATZ", "PSEXEC", "WMIC", "PROCDUMP", "RUBEUS", "SEATBELT",
"BLOODHOUND", "SHARPHOUND", "LAZAGNE", "SECRETSDUMP", "NTDSUTIL",
"CERTUTIL", "BITSADMIN", "MSHTA", "REGSVR32", "RUNDLL32",
"CSCRIPT", "WSCRIPT", "POWERSHELL", "CMD", "MSBUILD",
"INSTALLUTIL", "REGASM", "REGSVCS", "XWIZARD",
"NETCAT", "NCAT", "NC", "NMAP", "MASSCAN",
"RAR", "7Z", "WINRAR", "RCLONE",
]
def detect_suspicious_execution(prefetch_results):
"""Flag suspicious or known-attacker-tool prefetch files."""
findings = []
for result in prefetch_results:
name = (result.get("parsed_name") or "").upper()
for sus in SUSPICIOUS_EXECUTABLES:
if sus in name:
findings.append({
"severity": "HIGH",
"executable": result.get("parsed_name"),
"file": result.get("file"),
"reason": f"Known offensive/dual-use tool: {sus}",
"run_count": result.get("header", {}).get("run_count"),
"last_run": result.get("header", {}).get("last_run_time"),
})
break
return findings
def build_execution_timeline(prefetch_results):
"""Build a chronological timeline of program execution."""
timeline = []
for result in prefetch_results:
header = result.get("header", {})
last_run = header.get("last_run_time")
if last_run and last_run not in ("N/A", "Invalid timestamp"):
timeline.append({
"timestamp": last_run,
"executable": result.get("parsed_name"),
"run_count": header.get("run_count"),
"prefetch_file": result.get("file"),
})
return sorted(timeline, key=lambda x: x["timestamp"])
def run_pecmd(prefetch_path, output_dir=None):
"""Run Eric Zimmerman's PECmd for comprehensive prefetch parsing."""
import subprocess
cmd = ["PECmd.exe", "-f", prefetch_path]
if output_dir:
cmd += ["--csv", output_dir]
result = subprocess.run(cmd, capture_output=True, text=True, timeout=30)
return result.stdout, result.returncode
if __name__ == "__main__":
print("=" * 60)
print("Windows Prefetch File Analysis Agent")
print("Execution history, timeline building, suspicious tool detection")
print("=" * 60)
target = sys.argv[1] if len(sys.argv) > 1 else None
if target and os.path.exists(target):
if os.path.isdir(target):
print(f"\n[*] Scanning prefetch directory: {target}")
results = scan_prefetch_directory(target)
print(f"[*] Found {len(results)} prefetch files")
print("\n--- Execution History ---")
for r in results[:20]:
header = r.get("header", {})
name = r.get("parsed_name", "?")
count = header.get("run_count", "?")
last = header.get("last_run_time", "?")
print(f" {name:30s} runs={count} last={last}")
print("\n--- Suspicious Executables ---")
suspicious = detect_suspicious_execution(results)
for s in suspicious:
print(f" [!] {s['executable']}: {s['reason']} "
f"(runs={s['run_count']}, last={s['last_run']})")
print("\n--- Execution Timeline ---")
timeline = build_execution_timeline(results)
for t in timeline[-20:]:
print(f" {t['timestamp']} | {t['executable']} (x{t['run_count']})")
else:
print(f"\n[*] Analyzing: {target}")
exe_name, pf_hash = parse_prefetch_filename(target)
print(f" Name: {exe_name}, Hash: {pf_hash}")
header = parse_prefetch_header(target)
print(f" {json.dumps(header, indent=2)}")
else:
print(f"\n[DEMO] Usage:")
print(f" python agent.py <prefetch_dir> # Analyze all .pf files")
print(f" python agent.py <file.pf> # Analyze single prefetch file")