Pcapsummary

pip install cognis-pcapsummary
pcapsummary scan .            # → prioritized findings in seconds

Watch the walkthrough

A full narrated tour — setup, the tool in action, and every demo scenario:

Watch the pcapsummary walkthrough

▶ Watch the walkthrough (MP4)

🔎 Example output

Real, reproducible output from the tool — runs offline:

$ pcapsummary-emit --version
pcapsummary 0.1.0

$ pcapsummary-emit --help
usage: pcapsummary [-h] [--version] {summarize} ...

Summarize flows/talkers/protocols from a pcap text export (defensive analysis
only; no live capture).

positional arguments:
  {summarize}
    summarize  summarize a tshark-style text export

options:
  -h, --help   show this help message and exit
  --version    show program's version number and exit

Blocks above are real pcapsummary output — reproduce them from a clone.

Sample result format (illustrative values — run on your own data for real findings):

{
"findings": [
    {
        "id": "1234567890",
        "title": "Suspicious Network Traffic",
        "description": "Potential malicious activity detected on network interface 192.168.1.100",
        "created_at": "2023-02-20T14:30:00Z",
        "updated_at": "2023-02-20T14:30:01Z",
        "labels": ["malware", "network"],
        "observables": [
            {
                "type": "ip-dst",
                "value": "192.168.1.100"
            },
            {
                "type": "port",
                "value": 443
            }
        ]
    }
]
}

Usage — step by step

pcapsummary summarizes flows/talkers/protocols from a pcap text export (defensive analysis only; no live capture). Exit codes: 0 clean, 1 findings/parse errors, 2 usage/file error.

1. Install:

   pip install -e .
   pcapsummary --version
   

2. Produce a text export with tshark, then summarize it:

   tshark -r capture.pcap -T fields -e frame.time_relative -e ip.src -e ip.dst > export.txt
   pcapsummary summarize export.txt
   

3. Tune the report — cap top talkers/flows with --top, or read from stdin:

   pcapsummary summarize export.txt --top 20
   cat export.txt | pcapsummary summarize -
   

4. Read the output as JSON (protocols, talkers, flows, time span) or CSV (one row per flow — drops straight into a spreadsheet or pandas):

   pcapsummary summarize export.txt --format json | jq '.protocols'
   pcapsummary summarize export.txt --format csv  > flows.csv
   

5. Automate in CI — a non-zero exit flags parse errors or empty captures:

   pcapsummary summarize export.txt --format json > pcap.json || echo "parse issues"
   

Contents

• Why pcapsummary? · Features · Quick start · Example · Demos · Architecture · AI stack · How it compares · Integrations · Install anywhere · Related · Contributing

Why pcapsummary?

pcap at a glance

pcapsummary is single-purpose, scriptable, and self-hostable: point it at a target, get prioritized results in the format your workflow already speaks (table · JSON · SARIF), gate CI on it, and let agents drive it over MCP.

Features

• ✅ Parse comma- or tab-separated tshark/Wireshark text exports (header auto-detect)
• ✅ Summarize flows, top talkers (bidirectional), and protocol distribution
• ✅ Output as table, JSON, or CSV (one row per flow)
• ✅ Exit codes for CI gating (0 clean · 1 parse errors/empty · 2 usage)
• ✅ Eight worked demo scenarios (scan, beacon, exfil, DNS tunnel, lateral, …)
• ✅ Runs on Linux/macOS/Windows · Docker · devcontainer
• ✅ Ports in Python, JavaScript, Go, and Rust (ports/)

Quick start

pip install cognis-pcapsummary
pcapsummary --version
pcapsummary scan .                       # scan current project
pcapsummary scan . --format json         # machine-readable
pcapsummary scan . --fail-on high        # CI gate (non-zero exit)

Example

$ pcapsummary scan .
  [HIGH    ] PCA-001  example finding             (./src/app.py)
  [MEDIUM  ] PCA-002  another signal              (./config.yaml)

  2 findings · risk score 5 · 38ms

Demos

Five runnable, narrated scenarios live in demos/ — each runs the real pcapsummary API/CLI over a bundled offline capture export and explains the result for a specific audience. Every one exits 0 and is exercised by the test suite. Full write-up: docs/DEMOS.md.

Demo	Audience	What it shows
01_soc_triage	Network / SOC analyst	Top talkers, dominant protocol, heaviest flow — pcap at a glance
02_threat_hunter_scan	Threat hunter	One source fanning across 20 ports on one target = vertical TCP scan
03_ir_beacon_and_exfil	IR / forensics	Uniform C2 beacon cadence + upload-skewed byte asymmetry (exfil)
04_sysadmin_ci_gate	Sysadmin / DevOps	Distinct 0/1/2 exit codes a CI pipeline can gate on
05_dns_tunnel_and_lateral	Threat hunter / SOC	DNS dominance vs baseline + SMB sweep across 445 (lateral)

PYTHONUTF8=1 python demos/run_all.py                 # all five, end to end
PYTHONUTF8=1 python demos/02_threat_hunter_scan.py   # or just one

Each scenario reads the capture fixtures in the demos/NN-*/ folders, which also carry a SCENARIO.md documenting how the export was produced. Triage one directly with the CLI:

# Triage the scan capture as a table, then export its flows to CSV
python -m pcapsummary summarize demos/02-port-scan/capture_export.txt
python -m pcapsummary summarize demos/02-port-scan/capture_export.txt --format csv

All demos use private RFC 1918 addresses, RFC 5737 documentation ranges (203.0.113.0/24, 198.51.100.0/24), or real well-known public service IPs. They are static analysis of already-captured exports — no live capture, no network connections, authorized defensive use only.

Architecture

flowchart LR
  EXP["tshark / Wireshark<br/>text export"] --> PARSE["parse_export()<br/>dialect + header detect"]
  PARSE --> SUM["summarize()<br/>flows · talkers · protocols"]
  SUM --> OUT["table · JSON · CSV"]
  SUM --> MCP["MCP server<br/>(agents)"]

Full pipeline and data model: docs/ARCHITECTURE.md.

Use it from any AI stack

pcapsummary is interoperable with every popular way of using AI:

• MCP server — pcapsummary mcp (Claude Desktop, Cursor, Cognis.Studio, uncensored-fleet)
• OpenAI-compatible / JSON — pipe pcapsummary scan . --format json into any agent or LLM
• LangChain · CrewAI · AutoGen · LlamaIndex — wrap the CLI/JSON as a tool in one line
• CI / scripts — exit codes + SARIF for non-AI pipelines

How it compares

	Cognis pcapsummary	tshark
Self-hostable, no account	✅	varies
Single command, zero config	✅	⚠️
JSON + SARIF for CI	✅	varies
MCP-native (AI agents)	✅	❌
Polyglot ports (JS/Go/Rust)	✅	❌
Open license	✅ COCL	varies

Built in the spirit of tshark, re-framed the Cognis way. Missing a credit? Open a PR.

Integrations

Pipes into your stack: SARIF for code-scanning, JSON for anything, an MCP server (pcapsummary mcp) for AI agents, and a webhook forwarder for SIEM/Slack/Jira. See docs/INTEGRATIONS.md.

Install — every way, every platform

pip install "git+https://github.com/cognis-digital/pcapsummary.git"    # pip (works today)
pipx install "git+https://github.com/cognis-digital/pcapsummary.git"   # isolated CLI
uv tool install "git+https://github.com/cognis-digital/pcapsummary.git" # uv
pip install cognis-pcapsummary                                          # PyPI (when published)
docker run --rm ghcr.io/cognis-digital/pcapsummary:latest --help        # Docker
brew install cognis-digital/tap/pcapsummary                             # Homebrew tap
curl -fsSL https://raw.githubusercontent.com/cognis-digital/pcapsummary/main/install.sh | sh

Linux	macOS	Windows	Docker	Cloud
scripts/setup-linux.sh	scripts/setup-macos.sh	scripts/setup-windows.ps1	docker run ghcr.io/cognis-digital/pcapsummary	DEPLOY.md (AWS/Azure/GCP/k8s)

Related Cognis tools

• portfan — Summarize and diff nmap XML into prioritized, attackable findings
• subhunt — Aggregate & dedupe subdomain enumeration from multiple sources
• dirsight — Analyze web content-discovery output (ffuf/gobuster) into ranked endpoints
• jwtinspect — Decode JWTs and lint for alg=none, weak secrets, and missing claims
• corsaudit — Detect permissive/misconfigured CORS from headers or a config
• headerscan — Grade HTTP security headers (CSP/HSTS/XFO) A-F from a response dump

Explore the suite → 🗂️ all 170+ tools · ⭐ awesome-cognis · 🔗 cognis-sources · 🤖 uncensored-fleet · 🧠 engram

Contributing

PRs, new rules, and demo scenarios are welcome under the collaboration-pull model — see CONTRIBUTING.md and SECURITY.md.

⭐ If pcapsummary saved you time, star it — it genuinely helps others find it.

Interoperability

{} composes with the 300+ tool Cognis suite — JSON in/out and a shared OpenAI-compatible /v1 backbone. See INTEROP.md for the suite map, composition patterns, and reference stacks.

License

Source-available under the Cognis Open Collaboration License (COCL) v1.0 — free for personal, internal-evaluation, research, and educational use; commercial / production use requires a license ([email protected]). See LICENSE.

---