Extract and analyze Cobalt Strike beacon configuration from PE files and memory dumps to identify C2 infrastructure, malleable profiles, and operator tradecraft.
Extract and analyze Cobalt Strike beacon configuration from PE files and memory dumps to identify C2 infrastructure, malleable profiles, and operator tradecraft.
Cobalt Strike is a commercial adversary simulation tool widely abused by threat actors for post-exploitation operations. Beacon payloads contain embedded configuration data that reveals C2 server addresses, communication protocols, sleep intervals, jitter values, malleable C2 profile settings, watermark identifiers, and encryption keys. Extracting this configuration from PE files, shellcode, or memory dumps is critical for incident responders to map attacker infrastructure and attribute campaigns. The beacon configuration is XOR-encoded using a single byte (0x69 for version 3, 0x2e for version 4) and stored in a Type-Length-Value (TLV) format within the .data section.
When to Use
When investigating security incidents that require analyzing cobalt strike beacon configuration
When building detection rules or threat hunting queries for this domain
When SOC analysts need structured procedures for this analysis type
When validating security monitoring coverage for related attack techniques
Prerequisites
Python 3.9+ with dissect.cobaltstrike, pefile, yara-python
Network analysis tools (Wireshark) for C2 traffic correlation
Key Concepts
Beacon Configuration Structure
Cobalt Strike beacons store their configuration as a blob of TLV (Type-Length-Value) entries within the .data section of the PE. Stageless beacons XOR the entire beacon code with a 4-byte key. The configuration blob itself uses a single-byte XOR key. Each TLV entry contains a 2-byte type identifier (e.g., 0x0001 for BeaconType, 0x0008 for C2Server), a 2-byte length, and variable-length data.
Malleable C2 Profiles
The beacon configuration encodes the malleable C2 profile that dictates HTTP request/response transformations, including URI paths, headers, metadata encoding (Base64, NetBIOS), and data transforms. Analyzing these settings reveals how the beacon disguises its traffic to blend with legitimate web traffic.
Watermark and License Identification
Each Cobalt Strike license embeds a unique watermark (4-byte integer) into generated beacons. Extracting the watermark can link multiple beacons to the same operator or cracked license. Known watermark databases maintained by threat intelligence providers map watermarks to specific threat actors or leaked license keys.
Workflow
Step 1: Extract Configuration with CobaltStrikeParser
#!/usr/bin/env python3"""Extract Cobalt Strike beacon config from PE or memory dump."""import sys
import json
# Using SentinelOne's CobaltStrikeParser# pip install dissect.cobaltstrikefrom dissect.cobaltstrike.beacon import BeaconConfig
defextract_beacon_config(filepath):
"""Parse beacon configuration from file."""
configs = list(BeaconConfig.from_path(filepath))
ifnot configs:
print(f"[-] No beacon configuration found in {filepath}")
returnNonefor i, config inenumerate(configs):
print(f"\n[+] Beacon Configuration #{i+1}")
print(f"{'='*60}")
settings = config.as_dict()
# Critical fields for incident response
critical_fields = [
"SETTING_C2_REQUEST",
"SETTING_C2_RECOVER",
"SETTING_PUBKEY",
"SETTING_DOMAINS",
"SETTING_BEACONTYPE",
"SETTING_PORT",
"SETTING_SLEEPTIME",
"SETTING_JITTER",
"SETTING_MAXGET",
"SETTING_SPAWNTO_X86",
"SETTING_SPAWNTO_X64",
"SETTING_PIPENAME",
"SETTING_WATERMARK",
"SETTING_C2_VERB_GET",
"SETTING_C2_VERB_POST",
"SETTING_USERAGENT",
"SETTING_PROTOCOL",
]
for field in critical_fields:
value = settings.get(field, "N/A")
print(f" {field}: {value}")
return settings
returnNonedefextract_c2_indicators(config):
"""Extract actionable C2 indicators from beacon config."""
indicators = {
"c2_domains": [],
"c2_ips": [],
"c2_urls": [],
"user_agent": "",
"named_pipes": [],
"spawn_processes": [],
"watermark": "",
}
ifnot config:
return indicators
# Extract C2 domains
domains = config.get("SETTING_DOMAINS", "")
if domains:
for domain instr(domains).split(","):
domain = domain.strip().rstrip("/")
if domain:
indicators["c2_domains"].append(domain)
# Extract user agent
indicators["user_agent"] = str(config.get("SETTING_USERAGENT", ""))
# Extract named pipes
pipe = config.get("SETTING_PIPENAME", "")
if pipe:
indicators["named_pipes"].append(str(pipe))
# Extract spawn-to processesfor arch in ["SETTING_SPAWNTO_X86", "SETTING_SPAWNTO_X64"]:
proc = config.get(arch, "")
if proc:
indicators["spawn_processes"].append(str(proc))
# Extract watermark
indicators["watermark"] = str(config.get("SETTING_WATERMARK", ""))
return indicators
if __name__ == "__main__":
iflen(sys.argv) < 2:
print(f"Usage: {sys.argv[0]} <beacon_file_or_dump>")
sys.exit(1)
config = extract_beacon_config(sys.argv[1])
if config:
indicators = extract_c2_indicators(config)
print(f"\n[+] Extracted C2 Indicators:")
print(json.dumps(indicators, indent=2))
Step 2: Manual XOR Decryption of Beacon Config
import struct
deffind_and_decrypt_config(data):
"""Manually locate and decrypt beacon configuration."""# Cobalt Strike 4.x uses 0x2e as XOR key
xor_keys = [0x2e, 0x69] # v4, v3for xor_key in xor_keys:
# Search for the config magic bytes after XOR# Config starts with 0x0001 (BeaconType) XOR'd with key
magic = bytes([0x00 ^ xor_key, 0x01 ^ xor_key,
0x00 ^ xor_key, 0x02 ^ xor_key])
offset = data.find(magic)
if offset == -1:
continueprint(f"[+] Found config at offset 0x{offset:x} (XOR key: 0x{xor_key:02x})")
# Decrypt the config blob (typically 4096 bytes)
config_size = 4096
encrypted = data[offset:offset + config_size]
decrypted = bytes([b ^ xor_key for b in encrypted])
# Parse TLV entries
entries = parse_tlv(decrypted)
return entries
returnNonedefparse_tlv(data):
"""Parse Type-Length-Value configuration entries."""
entries = {}
offset = 0# TLV field type mapping
field_names = {
0x0001: "BeaconType",
0x0002: "Port",
0x0003: "SleepTime",
0x0004: "MaxGetSize",
0x0005: "Jitter",
0x0006: "MaxDNS",
0x0007: "Deprecated_PublicKey",
0x0008: "C2Server",
0x0009: "UserAgent",
0x000a: "PostURI",
0x000b: "Malleable_C2_Instructions",
0x000c: "Deprecated_HttpGet_Metadata",
0x000d: "SpawnTo_x86",
0x000e: "SpawnTo_x64",
0x000f: "CryptoScheme",
0x001a: "Watermark",
0x001d: "C2_HostHeader",
0x0024: "PipeName",
0x0025: "Year",
0x0026: "Month",
0x0027: "Day",
0x0036: "ProxyHostname",
}
while offset + 6 <= len(data):
entry_type = struct.unpack(">H", data[offset:offset+2])[0]
entry_len_type = struct.unpack(">H", data[offset+2:offset+4])[0]
entry_len = struct.unpack(">H", data[offset+4:offset+6])[0]
if entry_type == 0:
break
value_start = offset + 6
value_end = value_start + entry_len
value_data = data[value_start:value_end]
field_name = field_names.get(entry_type, f"Unknown_0x{entry_type:04x}")
if entry_len_type == 1: # Short
value = struct.unpack(">H", value_data[:2])[0]
elif entry_len_type == 2: # Int
value = struct.unpack(">I", value_data[:4])[0]
elif entry_len_type == 3: # String/Blob
value = value_data.rstrip(b'\x00').decode('utf-8', errors='replace')
else:
value = value_data.hex()
entries[field_name] = value
print(f" {field_name}: {value}")
offset = value_end
return entries