mirror of
https://github.com/juliandickie/claude-cybersecurity.git
synced 2026-08-24 17:47:44 +00:00
AI-powered cybersecurity code review skill for Claude Code. 8 specialist agents, OWASP 2025, CWE Top 25, MITRE ATT&CK, 11 languages, zero configuration.
https://agricidaniel.com/blog/claude-cybersecurity-ai-security-audit
- Shell 100%
bcc963894c
15-fix security remediation from self-audit (SECURITY-AUDIT-2026-04-15.md): - Remove WebFetch/WebSearch from allowed-tools (VULN-003/006) - Add prompt injection defense to agent dispatch (VULN-001) - Add per-agent tool restrictions for all 8 agents (VULN-010) - Add .claude/ to file exclusion list (VULN-004) - Add evidence redaction rules for secret masking (VULN-007) - Add Step 2.5 agent result validation (VULN-013) - Add integrity verification to install.sh (VULN-002) - Fix permission path mismatch in settings (VULN-014) - Remove nonexistent install.ps1 references (VULN-015) - Correct OWASP 2025 references to 2021 (VULN-011/012) - Remove unsubstantiated statistical claims (VULN-016/017) - Harden .gitignore with cert/key patterns (VULN-019) Marketplace compliance: - Fix plugin.json author format (string → object) - Add keywords and repository fields to plugin.json - Add plugin installation method to README - Align badges with actual capabilities (14 languages) New files: CHANGELOG.md, checksums.sha256, CI/CD pipeline, issue templates Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> |
||
|---|---|---|
| .claude-plugin | ||
| .github | ||
| assets | ||
| skills/cybersecurity | ||
| .gitignore | ||
| CHANGELOG.md | ||
| checksums.sha256 | ||
| CONTRIBUTING.md | ||
| install.sh | ||
| LICENSE | ||
| PRIVACY.md | ||
| README.md | ||
| research-prompt.md | ||
| SECURITY-AUDIT-2026-04-15.md | ||
| SECURITY.md | ||
| SUPPORT.md | ||
| uninstall.sh | ||
The most comprehensive AI-powered cybersecurity code review skill for Claude Code. Spawns 8 parallel specialist agents to audit your codebase across vulnerability detection, authorization verification, secret scanning, supply chain analysis, IaC security, threat intelligence (malware/C2/backdoor detection), AI-generated code patterns, and business logic flaws.
Complements GitHub Advanced Security by detecting what static tools architecturally cannot: missing security controls, business logic flaws, attack-path chaining, and obfuscated secrets — with zero configuration.
Installation
Manual (recommended)
git clone https://github.com/AgriciDaniel/claude-cybersecurity.git
cd claude-cybersecurity
bash install.sh
Plugin (Claude Code native)
claude plugin install cybersecurity
One-liner (convenience)
curl -fsSL https://raw.githubusercontent.com/AgriciDaniel/claude-cybersecurity/main/install.sh | bash
Demo
Quick Start
# Full security audit of current project
/cybersecurity
# Quick scan (entry points + auth + secrets + deps only)
/cybersecurity --scope quick
# Review only changed files (PR review mode)
/cybersecurity --scope diff
# Deep dive into one dimension
/cybersecurity --focus threat
# With compliance mapping
/cybersecurity --compliance pci
What It Does
Key Differentiators vs GitHub Advanced Security
| Capability | GHAS | This Skill |
|---|---|---|
| Business logic flaw detection | No | Yes |
| Authorization enforcement verification | Basic | Context-aware |
| Race condition detection | Very limited | Concurrency pattern analysis |
| Languages supported | 12 | 14 (pattern-based) + broader reasoning via LLM |
| IaC/Container/CI-CD scanning | No | Terraform, Docker, K8s, Actions |
| AI-generated code security | No | Specialized detection |
| Obfuscated secret detection | Regex only | Semantic context analysis |
| Threat intelligence (malware/C2) | No | MITRE ATT&CK mapped |
| Framework-aware false-positive suppression | No | 10 frameworks |
| Cost | $49/committer/month | Free (with Claude Code) |
Coverage
Scoring System
Architecture
File Structure
skills/cybersecurity/
├── SKILL.md (~990 lines — orchestrator)
├── references/
│ ├── vulnerability-taxonomy.md (25 CWE categories)
│ ├── scoring-rubric.md (formula + confidence system)
│ ├── threat-intelligence.md (MITRE ATT&CK patterns)
│ ├── compliance-matrix.md (5 frameworks)
│ ├── false-positive-suppression.md (10 frameworks)
│ ├── semgrep-patterns.md (8 detection patterns)
│ ├── report-template.md (output format + worked example)
│ ├── language-patterns/ (11 files)
│ └── iac-patterns/ (4 files)
Total: 23 files, 5,350 lines of security knowledge.
Requirements
- Claude Code (CLI, Desktop, or IDE extension)
- No other dependencies — zero configuration, works immediately
Uninstall
curl -fsSL https://raw.githubusercontent.com/AgriciDaniel/claude-cybersecurity/main/uninstall.sh | bash
Or manually:
rm -rf ~/.claude/skills/cybersecurity
Related Projects
- claude-seo — Comprehensive SEO analysis
- claude-blog — Full-lifecycle blog engine
- claude-ads — Paid advertising audit
License
MIT - AgriciDaniel 2026