
njones17/ai-agent-master-cyber-skills-list
Developer toolsThe most comprehensive cybersecurity skill pack for AI coding agents — 741 skills spanning offense, defense, cloud, forensics, malware analysis, CTF, AppSec, DevSecOps, OT/ICS, and more.
Overview
for AI coding agents — the most comprehensive single-repo collection available. Covers offense, defense, cloud, forensics, CTF, AppSec, DevSecOps, and more. Follows the agentskills.io open standard. Works with Claude, Claude Code, GitHub Copilot, OpenAI Codex, Cursor, Gemini CLI, and 20+ other platforms. Merged and deduplicated from 10 high-quality repos: See SOURCES.json for the full skill-level provenance map. - (same name: field): removed, kept best version - (different angle/depth): both kept - See DEDUP_REPORT.json for details Each skill follows the agentskills.io standard: Individual skills retain their original licenses. See each skill's directory for details. Collection structure: Apache 2.0.
README
Master Cybersecurity Skills
741 cybersecurity skills for AI coding agents — the most comprehensive single-repo collection available. Covers offense, defense, cloud, forensics, CTF, AppSec, DevSecOps, and more.
Follows the agentskills.io open standard. Works with Claude, Claude Code, GitHub Copilot, OpenAI Codex, Cursor, Gemini CLI, and 20+ other platforms.
Install
# Claude Code
/plugin marketplace add Njones17/AI-agent-master-cyber-skills-list
# Or clone directly
git clone https://github.com/Njones17/AI-agent-master-cyber-skills-list .skills/cybersecurity
Coverage
| Category | Skills |
|---|---|
| Cloud Security | AWS, Azure, GCP auditing, CSPM, IAM |
| Threat Intelligence | APT analysis, MITRE ATT&CK, STIX/TAXII, OSINT, dark web |
| Web Application Security | XSS, SQLi, SSRF, HTTP smuggling, cache poisoning |
| Offensive / Pentest | Metasploit, Burp Suite, recon, exploitation, C2 |
| Red Teaming | BloodHound, Kerberoasting, AD attacks, evasion |
| Digital Forensics | Volatility, disk imaging, memory, mobile, network |
| Malware Analysis | Ghidra, YARA, Cuckoo, static/dynamic analysis |
| SOC Operations | SIEM, Splunk, Windows Event Logs, detection rules |
| Network Security | Wireshark, Suricata, VLAN, IDS/IPS |
| Identity & Access | SAML, OAuth2, FIDO2, PAM, CyberArk, SailPoint |
| AppSec (SAST/DAST/SCA) | Semgrep, CodeQL, dependency scanning, secrets |
| Container Security | Trivy, Falco, Kubernetes hardening, Harbor |
| DevSecOps | GitLab CI, Gitleaks, supply chain, sbom |
| OT/ICS Security | SCADA, Modbus, Purdue Model |
| Cryptography | TLS, HSM, constant-time analysis, zeroize audit |
| CTF | Web, pwn, crypto, reverse engineering, forensics, OSINT |
| Incident Response | Ransomware, cloud IR, volatile evidence |
| Compliance | GDPR, ISO 27001, PCI DSS, CIS Benchmarks |
| AI/LLM Security | Prompt injection, model supply chain, red teaming |
| Smart Contracts | Solidity audit, timing side-channels, blockchain |
Sources
Merged and deduplicated from 10 high-quality repos:
| Repo | Stars | Contribution |
|---|---|---|
| trailofbits/skills | 3.3k | Security research, audit, static analysis, YARA, smart contracts |
| trailofbits/skills-curated | — | ffuf, Ghidra, Wooyun legacy, OpenAI security skills |
| ghostsecurity/skills | 356 | AppSec SAST/DAST/SCA/secrets scanning |
| itsmostafa/aws-agent-skills | 1k | AWS IAM, Secrets Manager, Cognito |
| mukul975/Anthropic-Cybersecurity-Skills | 9 | 611+ broad cybersecurity skills |
| narlyseorg/superhackers | — | Offensive pentest workflow (webapp/API/infra/Android) |
| ljagiello/ctf-skills | 18 | CTF: web, pwn, crypto, RE, forensics, OSINT, malware |
| BagelHole/DevOps-Security-Agent-Skills | 12 | 80+ DevOps/cloud security skills |
| Ray0907/security-scan | — | CVE/OWASP scanning with Semgrep |
| openai/skills | — | Security best practices, threat modeling, ownership map |
See SOURCES.json for the full skill-level provenance map.
Deduplication
- Exact duplicates (same
name:field): removed, kept best version - Similar skills (different angle/depth): both kept
- See
DEDUP_REPORT.jsonfor details
Structure
Each skill follows the agentskills.io standard:
skills/{skill-name}/
├── SKILL.md # YAML frontmatter + workflow instructions
├── references/ # Standards, MITRE ATT&CK, CVE refs (optional)
├── scripts/ # Helper scripts (optional)
└── assets/ # Templates, checklists (optional)
License
Individual skills retain their original licenses. See each skill’s directory for details. Collection structure: Apache 2.0.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add njones17/ai-agent-master-cyber-skills-list