Lists of “must-bookmark GitHub repos” are everywhere; what they rarely say is how the projects fit together. These ten cover two jobs. Six of them form a detection lab: watch the network and endpoints, write portable detection rules, and attack yourself to prove the rules fire. Four of them form an application-security pipeline: stop secrets, vulnerable dependencies and known weaknesses before code ships. For where each category sits in the commercial market, see Cybersecurity Tooling Landscape, Mapped.
1 · The Ten and the Job Each Does
| Project | What it does | Build | Licence |
|---|---|---|---|
| Wazuh | Endpoint agents plus a central server: log collection, file-integrity monitoring, vulnerability detection, alerting — a SIEM/XDR core | Lab | GPL-2.0 |
| Zeek | Network security monitor that turns traffic into rich logs (connections, DNS, HTTP, TLS, files) | Lab | BSD-3-Clause |
| Suricata | Signature-based intrusion detection and prevention engine | Lab | GPL-2.0 |
| Sigma | A generic rule format for log detections, convertible to many SIEM query languages; a large community rule set | Lab | Rules: Detection Rule License |
| YARA | Pattern-matching rules that identify and classify malware in files and memory | Lab | BSD-3-Clause |
| Atomic Red Team | A library of small, scripted attack techniques mapped to MITRE ATT&CK, for testing detections | Lab | MIT |
| Gitleaks | Scans Git repositories and commits for secrets: keys, tokens, passwords | Pipeline | MIT |
| Trivy | Scans container images, file systems, dependencies and infrastructure-as-code for vulnerabilities and misconfigurations; can produce SBOMs | Pipeline | Apache-2.0 |
| Nuclei | Template-driven scanner for known vulnerabilities and exposures in running services | Pipeline | MIT |
| OWASP Cheat Sheet Series | Concise, practical guidance on secure design and coding topics | Pipeline | CC BY-SA 4.0 |
Licences as published in each project’s repository at the time of writing; check before redistributing or embedding in a product. GPL projects can be used freely in-house; obligations arise when you distribute them.
2 · The Detection Lab
A lab that fits on one machine with a few virtual machines, or a small cloud account:
| Layer | Component | Set-up notes |
|---|---|---|
| Endpoints | Wazuh agents on a Windows and a Linux VM | Turn on Sysmon on Windows for process and network events; enable file-integrity monitoring on sensitive paths |
| Network | Zeek and Suricata on a sensor VM | Feed them a mirror (SPAN) port or virtual-switch mirror; ship their JSON logs to Wazuh |
| Central | Wazuh server with its indexer and dashboard | One place for alerts from both endpoints and network |
| Detection | Sigma rules | Convert chosen community rules to your back-end’s query language; keep your own rules in Git with review |
| Files | YARA rules | Scan downloads, mail attachments and memory images; Wazuh can trigger YARA scans on file changes |
▸ Start with ten Sigma rules you understand, not ten thousand you imported. A detection you cannot explain is one you cannot tune, and untuned rules bury real alerts.
3 · The AppSec Pipeline
Four checks, each placed where it is cheapest to fix what it finds:
| Stage | Tool | What it stops | Fail the build when… |
|---|---|---|---|
| Design & code review | OWASP Cheat Sheets | Known-bad patterns: injection, weak session handling, unsafe deserialisation | — (a reference for reviewers) |
| Commit | Gitleaks (pre-commit hook and CI) | Secrets entering the repository history | Any verified secret is found |
| Build | Trivy | Vulnerable dependencies and base images; misconfigured Dockerfiles, Kubernetes and Terraform | A fixable critical or a known-exploited CVE |
| Staging | Nuclei | Known exposures in the running service: default panels, misconfigurations, published CVEs | Any high-severity match on the staging host |
▸ A leaked key is not fixed by deleting the commit. Git history, forks and caches keep it. Rotate the secret first, then clean the history.
4 · Prove It Works: the Test Loop
A detection lab is only as good as the last time you tested it. Atomic Red Team turns testing into a routine:
- Pick a technique that matters to you from MITRE ATT&CK — for example credential dumping or a scheduled-task persistence.
- Run its atomic test on a lab VM. Read the test first: some change system settings or are destructive.
- Check the alert in Wazuh. Did a rule fire, with the right severity and enough context to act?
- Fix the gap — missing log source, a Sigma rule to add, or a noisy rule to tune.
- Record the result in a coverage sheet: technique, test, detected yes/no, date. Re-run after every change.
The coverage sheet becomes a metric worth reporting: share of priority ATT&CK techniques with a tested detection. See Cybersecurity KRIs.
5 · Ground Rules
| Rule | Why |
|---|---|
| Scan and attack only what you own or have written permission to test | In India, unauthorised access and damage to computer systems fall under sections 43 and 66 of the IT Act, 2000. Nuclei against someone else’s site is not research. See Cyber Crime & Law in India. |
| Keep the lab isolated | Atomic tests and malware samples must not reach production networks or real credentials. |
| Patch the tools | Security tools run with high privilege and parse hostile input; they are targets themselves. |
| Budget the people, not the licences | Open source moves cost from licences to engineering time: upgrades, tuning and on-call. |
Bottom line
- Ten projects, two builds — a detection lab and an AppSec pipeline.
- Wazuh is the hub; Zeek and Suricata see the network; Sigma and YARA are the portable rules.
- Atomic Red Team closes the loop — a detection you never tested is a guess.
- Shift left cheaply — Gitleaks at commit, Trivy at build, Nuclei in staging.
- Only test what you own, keep the lab isolated, and rotate any leaked secret first.
Sources & method. Capabilities and licences from each project’s official repository and documentation (Wazuh, Zeek, OISF Suricata, SigmaHQ, VirusTotal YARA, Red Canary Atomic Red Team, Gitleaks, Aqua Security Trivy, ProjectDiscovery Nuclei, OWASP). Repository popularity figures are deliberately omitted because they change daily. Architecture and pipeline placement are the author’s recommendations. Original analysis; no third-party graphics or text reproduced.
← Back to Cyber security