Agent skill

packet-capture

Expert skill for packet capture and analysis using libpcap/Wireshark. Execute tcpdump/tshark commands, write BPF filter expressions, analyze pcap files, decode protocol layers, calculate statistics, and generate Wireshark dissectors.

Stars 514
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/a5c-ai/babysitter/tree/main/library/specializations/network-programming/skills/packet-capture

Metadata

Additional technical details for this skill

author
babysitter-sdk
version
1.0.0
category
network-analysis
backlog id
SK-003

SKILL.md

packet-capture

You are packet-capture - a specialized skill for network packet capture and analysis, providing expert capabilities with libpcap, tcpdump, tshark, and Wireshark for deep network traffic inspection.

Overview

This skill enables AI-powered packet capture and analysis including:

  • Executing tcpdump/tshark commands and interpreting output
  • Writing and validating BPF filter expressions
  • Analyzing pcap/pcapng files
  • Decoding protocol layers (Ethernet, IP, TCP, UDP, application)
  • Calculating packet statistics and flow analysis
  • Generating Wireshark dissectors
  • Creating custom capture filters

Prerequisites

  • tcpdump or tshark installed
  • Root/admin privileges for live capture
  • Optional: Wireshark for GUI analysis
  • Optional: Python with scapy for programmatic analysis

Capabilities

1. Live Packet Capture

Capture network traffic with tcpdump and tshark:

bash
# Basic capture on interface
tcpdump -i eth0 -nn

# Capture with timestamp precision
tcpdump -i eth0 -nn -tttt

# Capture to file
tcpdump -i eth0 -w capture.pcap

# Capture with rotation (100MB files, keep 10)
tcpdump -i eth0 -w capture_%Y%m%d_%H%M%S.pcap -C 100 -W 10

# Capture specific traffic
tcpdump -i eth0 -nn 'port 80 or port 443'

# tshark capture with display filter
tshark -i eth0 -Y 'http.request.method == "GET"'

# tshark capture specific fields
tshark -i eth0 -T fields \
  -e frame.time \
  -e ip.src \
  -e ip.dst \
  -e tcp.port \
  -e http.host

2. BPF Filter Expressions

Write efficient Berkeley Packet Filter expressions:

bash
# Host filters
tcpdump host 192.168.1.100
tcpdump src host 192.168.1.100
tcpdump dst host 192.168.1.100

# Network filters
tcpdump net 192.168.1.0/24
tcpdump src net 10.0.0.0/8

# Port filters
tcpdump port 80
tcpdump src port 443
tcpdump portrange 8000-8100

# Protocol filters
tcpdump tcp
tcpdump udp
tcpdump icmp
tcpdump 'ip proto 47'  # GRE

# Combining filters
tcpdump 'host 192.168.1.100 and port 80'
tcpdump 'src host 192.168.1.100 or dst host 192.168.1.100'
tcpdump 'tcp and (port 80 or port 443)'
tcpdump 'not port 22'  # Exclude SSH

# TCP flag filters
tcpdump 'tcp[tcpflags] & (tcp-syn|tcp-fin) != 0'
tcpdump 'tcp[tcpflags] & tcp-syn != 0'  # SYN packets
tcpdump 'tcp[tcpflags] & tcp-rst != 0'  # RST packets

# Payload filters
tcpdump 'tcp[((tcp[12:1] & 0xf0) >> 2):4] = 0x47455420'  # HTTP GET
tcpdump 'tcp[((tcp[12:1] & 0xf0) >> 2):4] = 0x504f5354'  # HTTP POST

# VLAN filters
tcpdump 'vlan 100'
tcpdump 'vlan and host 192.168.1.100'

3. PCAP File Analysis

Analyze captured packet files:

bash
# Read pcap file
tcpdump -r capture.pcap -nn

# Count packets
tcpdump -r capture.pcap | wc -l

# Extract specific fields with tshark
tshark -r capture.pcap -T fields \
  -e frame.number \
  -e frame.time_relative \
  -e ip.src \
  -e ip.dst \
  -e tcp.stream \
  -e http.request.uri

# Protocol hierarchy statistics
tshark -r capture.pcap -q -z io,phs

# Conversation statistics
tshark -r capture.pcap -q -z conv,tcp
tshark -r capture.pcap -q -z conv,ip

# Endpoint statistics
tshark -r capture.pcap -q -z endpoints,tcp

# HTTP statistics
tshark -r capture.pcap -q -z http,tree
tshark -r capture.pcap -q -z http_req,tree

# Follow TCP stream
tshark -r capture.pcap -q -z follow,tcp,ascii,0

# Export objects (HTTP, SMB, etc.)
tshark -r capture.pcap --export-objects http,./http_exports/

# Time-based statistics
tshark -r capture.pcap -q -z io,stat,1

4. Protocol Layer Decoding

Decode and analyze protocol layers:

python
from scapy.all import *

def analyze_packet(packet):
    """Analyze packet layers."""
    analysis = {}

    # Ethernet layer
    if Ether in packet:
        eth = packet[Ether]
        analysis['ethernet'] = {
            'src': eth.src,
            'dst': eth.dst,
            'type': hex(eth.type)
        }

    # IP layer
    if IP in packet:
        ip = packet[IP]
        analysis['ip'] = {
            'version': ip.version,
            'ihl': ip.ihl,
            'tos': ip.tos,
            'len': ip.len,
            'id': ip.id,
            'flags': str(ip.flags),
            'frag': ip.frag,
            'ttl': ip.ttl,
            'proto': ip.proto,
            'src': ip.src,
            'dst': ip.dst
        }

    # TCP layer
    if TCP in packet:
        tcp = packet[TCP]
        analysis['tcp'] = {
            'sport': tcp.sport,
            'dport': tcp.dport,
            'seq': tcp.seq,
            'ack': tcp.ack,
            'flags': str(tcp.flags),
            'window': tcp.window,
            'options': [(name, val) for name, val in tcp.options]
        }

    # UDP layer
    if UDP in packet:
        udp = packet[UDP]
        analysis['udp'] = {
            'sport': udp.sport,
            'dport': udp.dport,
            'len': udp.len
        }

    # HTTP layer (raw)
    if Raw in packet:
        payload = packet[Raw].load
        if payload.startswith(b'GET ') or payload.startswith(b'POST ') or \
           payload.startswith(b'HTTP/'):
            analysis['http'] = {
                'payload': payload[:500].decode('utf-8', errors='replace')
            }

    return analysis

def analyze_pcap(filename):
    """Analyze all packets in a pcap file."""
    packets = rdpcap(filename)
    results = []

    for i, pkt in enumerate(packets):
        result = {
            'number': i + 1,
            'time': float(pkt.time),
            'length': len(pkt),
            'layers': analyze_packet(pkt)
        }
        results.append(result)

    return results

5. Flow Analysis

Analyze network flows and connections:

python
from collections import defaultdict
from scapy.all import *

def extract_flows(pcap_file):
    """Extract TCP/UDP flows from pcap file."""
    packets = rdpcap(pcap_file)
    flows = defaultdict(lambda: {
        'packets': [],
        'bytes': 0,
        'start_time': None,
        'end_time': None
    })

    for pkt in packets:
        if IP not in pkt:
            continue

        src_ip = pkt[IP].src
        dst_ip = pkt[IP].dst

        if TCP in pkt:
            src_port = pkt[TCP].sport
            dst_port = pkt[TCP].dport
            proto = 'tcp'
        elif UDP in pkt:
            src_port = pkt[UDP].sport
            dst_port = pkt[UDP].dport
            proto = 'udp'
        else:
            continue

        # Create bidirectional flow key
        if (src_ip, src_port) < (dst_ip, dst_port):
            flow_key = (src_ip, src_port, dst_ip, dst_port, proto)
        else:
            flow_key = (dst_ip, dst_port, src_ip, src_port, proto)

        flow = flows[flow_key]
        flow['packets'].append(pkt)
        flow['bytes'] += len(pkt)

        pkt_time = float(pkt.time)
        if flow['start_time'] is None or pkt_time < flow['start_time']:
            flow['start_time'] = pkt_time
        if flow['end_time'] is None or pkt_time > flow['end_time']:
            flow['end_time'] = pkt_time

    return flows

def flow_statistics(flows):
    """Calculate statistics for flows."""
    stats = []
    for key, flow in flows.items():
        src_ip, src_port, dst_ip, dst_port, proto = key
        duration = flow['end_time'] - flow['start_time'] if flow['end_time'] != flow['start_time'] else 0.001

        stats.append({
            'src': f"{src_ip}:{src_port}",
            'dst': f"{dst_ip}:{dst_port}",
            'protocol': proto,
            'packets': len(flow['packets']),
            'bytes': flow['bytes'],
            'duration': duration,
            'bps': flow['bytes'] * 8 / duration,
            'pps': len(flow['packets']) / duration
        })

    return sorted(stats, key=lambda x: x['bytes'], reverse=True)

6. Wireshark Dissector Generation

Generate custom Wireshark dissectors:

lua
-- Custom Protocol Dissector for Lua
-- Save as myprotocol.lua in Wireshark plugins directory

local myproto = Proto("myproto", "My Custom Protocol")

-- Define fields
local f_magic = ProtoField.uint8("myproto.magic", "Magic", base.HEX)
local f_version = ProtoField.uint8("myproto.version", "Version", base.DEC)
local f_type = ProtoField.uint8("myproto.type", "Message Type", base.DEC)
local f_flags = ProtoField.uint8("myproto.flags", "Flags", base.HEX)
local f_length = ProtoField.uint32("myproto.length", "Payload Length", base.DEC)
local f_payload = ProtoField.bytes("myproto.payload", "Payload")

myproto.fields = { f_magic, f_version, f_type, f_flags, f_length, f_payload }

-- Message type names
local type_names = {
    [0x01] = "HANDSHAKE",
    [0x02] = "DATA",
    [0x03] = "ACK",
    [0x04] = "ERROR",
    [0x05] = "CLOSE"
}

-- Dissector function
function myproto.dissector(buffer, pinfo, tree)
    local length = buffer:len()
    if length < 8 then return end  -- Minimum header size

    pinfo.cols.protocol = myproto.name

    local subtree = tree:add(myproto, buffer(), "My Protocol Data")

    -- Parse header
    local magic = buffer(0, 1):uint()
    if magic ~= 0xAB then return 0 end  -- Not our protocol

    subtree:add(f_magic, buffer(0, 1))
    subtree:add(f_version, buffer(1, 1))

    local msg_type = buffer(2, 1):uint()
    local type_item = subtree:add(f_type, buffer(2, 1))
    type_item:append_text(" (" .. (type_names[msg_type] or "UNKNOWN") .. ")")

    subtree:add(f_flags, buffer(3, 1))

    local payload_len = buffer(4, 4):uint()
    subtree:add(f_length, buffer(4, 4))

    -- Parse payload if present
    if payload_len > 0 and length >= 8 + payload_len then
        subtree:add(f_payload, buffer(8, payload_len))
    end

    -- Update info column
    pinfo.cols.info = type_names[msg_type] or "Unknown"

    return 8 + payload_len
end

-- Register on TCP port
local tcp_port = DissectorTable.get("tcp.port")
tcp_port:add(12345, myproto)

MCP Server Integration

This skill can leverage the following MCP servers for enhanced capabilities:

Server Description Integration
Wireshark MCP (sarthaksiddha) Live capture and analysis AI-powered packet inspection
WireMCP Real-time traffic analysis Threat detection
mcp-wireshark Wireshark/tshark integration IDE integration
Network Monitor MCP Security analysis Real-time monitoring

Wireshark MCP Server

bash
# Install from npm
npm install -g @sarthaksiddha/wireshark-mcp

# Add to Claude
claude mcp add wireshark -- npx @sarthaksiddha/wireshark-mcp

Capabilities:

  • Live traffic capture
  • PCAP file analysis
  • Protocol statistics
  • TCP stream following
  • JSON export

Best Practices

  1. Use capture filters - Reduce capture overhead at kernel level
  2. Rotate capture files - Prevent disk exhaustion
  3. Set snap length - Capture only needed bytes with -s
  4. Use ring buffers - For continuous capture with -W
  5. Filter early - BPF filters are more efficient than display filters
  6. Anonymize data - Remove sensitive information before sharing

Process Integration

This skill integrates with the following processes:

  • packet-capture-analysis.js - Packet capture and analysis
  • protocol-dissector.js - Protocol dissection
  • network-traffic-analyzer.js - Traffic pattern analysis

Output Format

When executing operations, provide structured output:

json
{
  "operation": "analyze",
  "file": "capture.pcap",
  "status": "success",
  "summary": {
    "packets": 10000,
    "bytes": 5242880,
    "duration": 60.5,
    "avgPacketSize": 524
  },
  "protocols": {
    "tcp": 8500,
    "udp": 1200,
    "icmp": 300
  },
  "topTalkers": [
    {"ip": "192.168.1.100", "packets": 3000, "bytes": 1500000}
  ],
  "flows": 150,
  "artifacts": ["analysis.json", "flows.csv"]
}

Constraints

  • Require appropriate permissions for live capture
  • Respect privacy and legal requirements
  • Limit capture duration and size
  • Filter sensitive protocols (credentials, PII)
  • Store captures securely

Expand your agent's capabilities with these related and highly-rated skills.

a5c-ai/babysitter

gsd-tools

Central utility skill for GSD operations. Provides config parsing, slug generation, timestamps, path operations, and orchestrates calls to other specialized skills. Acts as the unified entry point that the original gsd-tools.cjs provided via its lib/ modules (commands, config, core, init).

514 31
Explore
a5c-ai/babysitter

model-profile-resolution

Resolve model profile (quality/balanced/budget) at orchestration start and map agents to specific models. Enables cost/quality tradeoffs by selecting appropriate AI models for each agent role.

514 31
Explore
a5c-ai/babysitter

verification-suite

Plan structure validation, phase completeness checks, reference integrity verification, and artifact existence confirmation. Provides the structured verification layer ensuring GSD artifacts are well-formed and complete.

514 31
Explore
a5c-ai/babysitter

state-management

STATE.md reading, writing, and field-level updates. Provides cross-session state persistence via .planning/STATE.md with structured fields for current task, completed phases, blockers, decisions, and quick tasks.

514 31
Explore
a5c-ai/babysitter

git-integration

Git commit patterns, formats, and conventions for GSD methodology. Provides atomic commits per task, structured commit messages, planning file commits, branch management, and milestone tag operations.

514 31
Explore
a5c-ai/babysitter

frontmatter-parsing

YAML frontmatter parsing and manipulation for .planning/ documents. Provides read, write, update, query, and validation operations on frontmatter blocks in GSD markdown artifacts.

514 31
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results