Metascrub

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

🔎 Example output

Real, reproducible output from the tool — runs offline:

$ metascrub-emit --version
metascrub 0.1.0

$ metascrub-emit --help
usage: metascrub [-h] [--version] [--format {table,json}] {scan,clean} ...

Strip identifying metadata from documents and images before release
(PNG/JPEG/PDF). Defensive use only.

positional arguments:
  {scan,clean}
    scan                report metadata without modifying files
    clean               write sanitized copies with metadata removed

options:
  -h, --help            show this help message and exit
  --version             show program's version number and exit
  --format {table,json}
                        output format (default: table)

Blocks above are real metascrub 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": "Anomalous network traffic detected from IP 192.168.1.100",
        "created_by": "[email protected]",
        "created_at": "2022-07-15T14:30:00Z"
    }
]
}

Usage — step by step

1. Install (Python 3.8+, stdlib only):

   pip install metascrub        # or: pipx install metascrub
   

2. Scan a file for identifying metadata (read-only, no changes written):

   metascrub scan report.pdf photo.jpg
   

   scan exits 1 when any metadata is found, 0 when clean — so it doubles as a gate.
3. Write a sanitized copy with metadata stripped:

   metascrub clean photo.jpg -o photo.clean.jpg     # single file -> explicit output
   metascrub clean *.png --in-place                 # overwrite in place
   

   (-o/--output is only valid with a single input file.)
4. Read the output as JSON for tooling / dashboards:

   metascrub --format json scan report.pdf | jq '.total_findings, .results[].findings'
   

   The payload includes tool, version, action, results[], total_findings, and errors.
5. Gate a release pipeline in CI — fail the build if any artifact still carries metadata:

   metascrub scan dist/*.pdf dist/*.png || { echo "metadata leak — blocking release"; exit 1; }
   

Contents

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

Why metascrub?

clean before ship

metascrub 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

• ✅ Detect Format
• ✅ Scan File
• ✅ Clean File
• ✅ Runs on Linux/macOS/Windows · Docker · devcontainer
• ✅ Ports in Python, JavaScript, Go, and Rust (ports/)

Quick start

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

Example

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

  2 findings · risk score 5 · 38ms

Architecture

flowchart LR
  IN[image / coordinates] --> P[metascrub<br/>extract + geolocate]
  P --> OUT[location estimate]

Use it from any AI stack

metascrub is interoperable with every popular way of using AI:

• MCP server — metascrub mcp (Claude Desktop, Cursor, Cognis.Studio, uncensored-fleet)
• OpenAI-compatible / JSON — pipe metascrub 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 metascrub	mat2
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 mat2, 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 (metascrub 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/metascrub.git"    # pip (works today)
pipx install "git+https://github.com/cognis-digital/metascrub.git"   # isolated CLI
uv tool install "git+https://github.com/cognis-digital/metascrub.git" # uv
pip install cognis-metascrub                                          # PyPI (when published)
docker run --rm ghcr.io/cognis-digital/metascrub:latest --help        # Docker
brew install cognis-digital/tap/metascrub                             # Homebrew tap
curl -fsSL https://raw.githubusercontent.com/cognis-digital/metascrub/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/metascrub	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 metascrub 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.

---