Network capture analysis with scapy: stream extraction, DNS/HTTP parsing, credential detection, anomaly detection, file carving

GitHub stars
18

Install

/plugin install pcap-analyzer@my-claude-skills

Run it inside Claude Code after adding the marketplace with /plugin marketplace add sandbornm/my-claude-skills.

Contents

Discovered from source

Skills 1

Pcap Network Capture Analysis

Perform automated network traffic analysis on pcap/pcapng files using scapy. Extract streams, DNS, HTTP, credentials, and files. Detect anomalies like beaconing, port scanning, C2 patterns, and data exfiltration.

When to Use

  • Analyzing packet captures from incident response or forensic investigations
  • Extracting network artifacts (files, credentials, DNS queries) from pcaps
  • Detecting C2 beaconing, port scanning, or data exfiltration patterns
  • Reconstructing HTTP sessions and downloaded files from captures
  • Mapping network endpoints and communication patterns
  • Analyzing malware network behavior from sandbox captures
  • CTF challenges involving network forensics
  • Investigating cleartext credential exposure in network traffic

When NOT to Use

  • Live traffic capture — use tcpdump, Wireshark, or tshark directly
  • Need Wireshark GUI features — open the pcap in Wireshark
  • Analyzing encrypted TLS content — need SSLKEYLOGFILE and tshark
  • NetFlow/IPFIX data — use flow analysis tools (nfdump, SiLK)
  • Wireless-specific analysis (802.11) — use Aircrack-ng suite
  • IDS/IPS rule matching — use Suricata or Snort

Quick Reference

TaskCommand
Full analysis{baseDir}/scripts/pcap-analyze.sh -s analyze_all.py -o ./output capture.pcap
Extract TCP/UDP streams{baseDir}/scripts/pcap-analyze.sh -s extract_streams.py -o ./output capture.pcap
Extract DNS queries{baseDir}/scripts/pcap-analyze.sh -s extract_dns.py -o ./output capture.pcap
Extract HTTP traffic{baseDir}/scripts/pcap-analyze.sh -s extract_http.py -o ./output capture.pcap
Map endpoints{baseDir}/scripts/pcap-analyze.sh -s export_endpoints.py -o ./output capture.pcap
Traffic statistics{baseDir}/scripts/pcap-analyze.sh -s export_statistics.py -o ./output capture.pcap
Detect anomalies{baseDir}/scripts/pcap-analyze.sh -s find_anomalies.py -o ./output capture.pcap
Find credentials{baseDir}/scripts/pcap-analyze.sh -s extract_credentials.py -o ./output capture.pcap
Carve files{baseDir}/scripts/pcap-analyze.sh -s extract_files.py -o ./output capture.pcap

Prerequisites

  • Python 3.8+ with pip
  • scapy: pip install scapy
  • tshark (optional, recommended): Install Wireshark (brew install wireshark / apt install tshark)

Main Wrapper Script

{baseDir}/scripts/pcap-analyze.sh [options] <capture_file>

Options:

  • -o, --output <dir> — Output directory for results (default: current dir)
  • -s, --script <name> — Analysis script to run (can be repeated)
  • -a, --script-args <args> — Arguments for the last specified script
  • --bpf <filter> — BPF filter to apply before analysis
  • --timeout <seconds> — Analysis timeout
  • -v, --verbose — Verbose output
  • -h, --help — Show help

Built-in Analysis Scripts

analyze_all.py

Comprehensive first-pass analysis. Best for initial triage of unknown captures.

Output files:

  • {name}_summary.txt — Capture overview: packet count, time range, duration, data rates
  • {name}_protocols.json — Protocol distribution by layer
  • {name}_endpoints.json — Top talkers by packet and byte count
  • {name}_conversations.json — Top IP-pair conversations
  • {name}_interesting.txt — Notable findings categorized as cleartext protocols, unusual ports, large transfers, suspicious DNS, tunneling indicators

extract_streams.py

Reassemble TCP and UDP streams. Pass --host <ip> or --port <port> to filter, --max-streams <n> to limit output.

Output: {name}_streams.json + {name}_stream_{n}.bin per stream

extract_dns.py

Extract all DNS queries and responses with full record detail.

Output:

  • {name}_dns.json — Queries, responses, record types, TTLs
  • {name}_domains.txt — Unique domains queried
  • {name}_dns_timeline.json — Temporal query analysis

extract_http.py

Extract HTTP transactions from raw TCP payloads. Handles chunked transfer encoding.

Output: {name}_http.json + {name}_http_bodies/ directory

export_endpoints.py

Map all network endpoints with service identification. Includes MAC addresses, ports, protocols, subnets, and connection 5-tuples.

Output: {name}_endpoints.json

export_statistics.py

Detailed traffic statistics: protocol hierarchy, packet size distribution, timing analysis, TCP flag distribution, TTL distribution with OS fingerprint hints.

Output: {name}_statistics.json

find_anomalies.py

Detect suspicious patterns: C2 beaconing (low coefficient of variation in connection intervals), port scanning, data exfiltration, DNS tunneling (high-entropy long subdomain labels), unusual protocol usage, cleartext sensitive data, and TLS anomalies.

Output: {name}_anomalies.json

extract_credentials.py

Find cleartext credentials: HTTP Basic/Digest auth, form POST data, FTP USER/PASS, SMTP AUTH, Telnet logins, session cookies, API keys. Passwords are partially redacted in output.

Output: {name}_credentials.json

extract_files.py

Carve files from HTTP responses, FTP transfers, SMTP attachments, and magic-byte detection in TCP streams. Computes MD5 hashes. Supports PDF, ZIP, PNG, JPEG, PE, ELF, GIF, GZIP, and more.

Output: {name}_files.json + {name}_extracted/ directory

Common Workflows

Triage an Unknown Capture

mkdir -p ./analysis
{baseDir}/scripts/pcap-analyze.sh -s analyze_all.py -o ./analysis capture.pcap
cat ./analysis/capture_summary.txt
cat ./analysis/capture_interesting.txt

Investigate Suspected C2 Traffic

{baseDir}/scripts/pcap-analyze.sh -s find_anomalies.py -o ./c2 capture.pcap
{baseDir}/scripts/pcap-analyze.sh -s extract_dns.py -o ./c2 capture.pcap
cat ./c2/capture_anomalies.json | jq '.beaconing'
cat ./c2/capture_anomalies.json | jq '.data_exfiltration'

Extract Files from HTTP Traffic

{baseDir}/scripts/pcap-analyze.sh -s extract_http.py -s extract_files.py -o ./files capture.pcap
cat ./files/capture_files.json | jq '.[].filename'
ls ./files/capture_extracted/

Credential Exposure Audit

{baseDir}/scripts/pcap-analyze.sh -s extract_credentials.py -o ./creds capture.pcap
cat ./creds/capture_credentials.json | jq 'to_entries | map(select(.value | length > 0))'

Network Forensics (Full Workflow)

mkdir -p ./forensics
{baseDir}/scripts/pcap-analyze.sh \
    -s analyze_all.py \
    -s export_endpoints.py \
    -s extract_dns.py \
    -s find_anomalies.py \
    -s extract_credentials.py \
    -s extract_files.py \
    -o ./forensics evidence.pcap

Filter Analysis to Specific Host

{baseDir}/scripts/pcap-analyze.sh \
    --bpf "host 192.168.1.100" \
    -s analyze_all.py \
    -o ./host_analysis capture.pcap

{baseDir}/scripts/pcap-analyze.sh \
    -s extract_streams.py \
    -a "--host 192.168.1.100" \
    -o ./host_analysis capture.pcap

Troubleshooting

scapy Not Installed

pip install scapy

Large Capture Files

For captures over 100MB, use BPF filters to narrow scope:

{baseDir}/scripts/pcap-analyze.sh --bpf "tcp port 80" -s extract_http.py capture.pcap

Or use individual scripts instead of analyze_all.py.

Permission Errors Reading Pcap

chmod 644 capture.pcap
# Or run with sudo if needed

tshark Not Found

tshark is optional. All scripts work with scapy alone. For best results:

# macOS
brew install wireshark

# Debian/Ubuntu
sudo apt install tshark

# Fedora/RHEL
sudo dnf install wireshark-cli

Tips

  1. Start with analyze_all.py — gives a full overview for triage
  2. BPF filters reduce noise — use --bpf to focus on specific hosts or protocols
  3. Use jq for JSON — all JSON exports are designed for machine processing
  4. Combine with angr/Frida — extract a binary from pcap, then analyze with angr or trace with Frida
  5. Large pcaps — filter first, analyze second; scapy loads entire capture into memory
  6. Credentials are redacted — extract_credentials.py partially masks passwords for safe reporting

Sponsors

Details

Marketplace
sandbornm/my-claude-skills
Source
pcap-analyzer
Category
Community

More from sandbornm/my-claude-skills

All 5