AI tools, scored for your job
Get paid to train AI
🤝 You're in the Free & Open Source library — community-built tools you can use without paying. A different concept from our scored AI directory.
Home › Free & Open Source › Compliance › Intrusion-and-anomaly-detection-with-machine-learning
✦ Free & open sourceCompliance

Intrusion-and-anomaly-detection-with-machine-learning

Detects attacks in application logs using DBSCAN clustering, PCA, and local LLM analysis.

Unsupervised AI log anomaly detection and incident analysis

Cluster HTTP and system logs using PCA and DBSCAN, then analyze outliers with local Ollama LLMs.

Free forever · open source · MIT
✓ Needs IT✓ Browser-based✓ Real-time collaboration
6.6
ZEKAI
Unsupervised AI log security
Free & open source · Compliance
Ease of setup4
Usefulness8
Maintenance9.5
Community5
★ 175
GITHUB
Oct 2026
UPDATED
MIT
LICENSE
🛠 Don't want to self-host? We can set it up →
⚡ Quick answer

Webhawk (Catch 3.0) is an open-source security tool that identifies attack traces in web server and system logs without pre-written rules. It applies unsupervised machine learning to cluster log entries, flagging anomalous activity for automated LLM review. SOC analysts and compliance teams can use it to surface hidden threats across endpoints.

Runs on
Web
Hosting
Self-host
Languages
Ruby
License
MIT

What it is

The engine converts raw logs into numerical features via PCA and applies DBSCAN density clustering to isolate outliers. Anomalous logs pass to a local LLM running via Ollama to generate incident descriptions and cross-reference known CVEs. Detections stream directly to a web interface or integrate with existing SIEM environments through a REST API.

Who it's for

It suits security analysts, incident responders, and GRC teams looking to catch zero-day attack patterns across Apache, Nginx, or process logs. It is not suitable for non-technical teams or environments lacking dedicated compute for self-hosted AI models.

Key features

Unsupervised DBSCAN Clustering

Groups routine log entries by feature density and flags outliers as potential attacks without signature rules.

Local LLM Analysis

Routes detected anomalies to Ollama models to generate context-aware incident summaries and advice.

Automated CVE Mapping

Queries the NVD database to correlate detected log anomalies with published security vulnerabilities.

Endpoint Log Collector

Provides a Python agent to read local access log files and ship them to the central detection engine.

How to start

Up and running in 3 steps

1
Clone the repository and copy the settings template files for the engine, agent, and web application.
2
Launch the detection API, web UI, and Ollama model services using Docker Compose.
3
Deploy the Python log agent on target endpoints or execute the catch script against static log files.
Real talk

Our honest verdict

What's great

  • Identifies novel attack patterns without requiring pre-configured signature rules.
  • Runs entirely on your infrastructure with local Ollama models to protect log privacy.
  • Combines PCA feature extraction with configurable DBSCAN hyperparameter tuning.
  • Integrates with existing SOC setups using built-in REST API endpoints.

Where it falls short

  • Requires server resources to host Docker containers and run local LLM inference.
  • Python agents must be manually installed and configured on every monitored endpoint.
  • Configuring DBSCAN distance metrics requires statistical or machine learning familiarity to control false positive rates.
Cost

Pricing & license

Webhawk is free and open-source under the MIT license. The primary cost comes from providing the server hardware required to host the Docker stack and run local LLMs.

Paid work

Open remote AI-training roles

877 open remote AI-training roles, updated November 16, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.

See all open roles →
Compare

Alternatives

🎁 Free & open-source alternatives

These reviewed AI tools do the same job instantly — from our scored directory.

Get more

Use it, or have us set it up

📄 Free PDF: Intrusion-and-anomaly-detection-with-machine-learning for Compliance
A plain-language guide to getting your first result. Sent to your inbox.
🔔 64 people want a full video tutorial — add your vote
🛠 Don't want to set it up yourself?
We'll install, host & configure Intrusion-and-anomaly-detection-with-machine-learning for you — and train your team.
Get it done for you ↗
Questions

FAQ

What log formats does Webhawk support out of the box?

It includes built-in parsing regex patterns for Apache access logs, Nginx access logs, HTTP logs, and OS process details.

Does Webhawk send log data to external cloud APIs?

No, default configurations route log analysis to a self-hosted Ollama instance running local models like Granite4.

Radar

Latest AI news

✉️ The weekly free toolkit
The best new free & open-source tools for Compliance — one short email a week, in your language.
Free · 1 email/week · unsubscribe anytime
Get paid to train AI →

See Zekai first in Google