Cybercentry Verification — Claude Code Skill

A verification-first OWASP security review skill for Claude Code. Threat-model driven, risk-prioritized, and anchored to the full OWASP standards suite — Top 10:2025, ASVS 5.0, API Security Top 10, and the Top 10 for LLM Applications 2025.

Cybercentry: AI Agent Cyber Security | Verification | Risk-Aware Systems.

What it does

Turns Claude into Cybercentry — a sharp security verifier that:

• Threat-models first — maps assets, entry points, trust boundaries, and data flows before reviewing code
• Anchors every finding to OWASP — Top 10:2025, ASVS 5.0, API Security, LLM Top 10 2025
• Prioritizes by risk — Critical and High first, with business/exploit impact
• Delivers surgical fixes — concrete diffs, config snippets, and exact tool commands (Semgrep, ZAP, Snyk, Trivy)
• Covers AI/agent-specific risks — prompt injection, supply chain, deserialization, vector/embedding weaknesses

Findings come back in a consistent format:

OWASP Ref | Severity | Finding | Location | Impact | Verification Steps | Recommended Fix

Install

Option 1 — Global install (recommended)

Available across every project on your machine.

macOS / Linux:

git clone https://github.com/Cybercentry/cybercentry-verification.git
mkdir -p ~/.claude/skills
cp -r cybercentry-verification/skills/cybercentry-verification ~/.claude/skills/

Windows (PowerShell):

git clone https://github.com/Cybercentry/cybercentry-verification.git
New-Item -ItemType Directory -Force -Path "$HOME\.claude\skills" | Out-Null
Copy-Item -Recurse cybercentry-verification\skills\cybercentry-verification "$HOME\.claude\skills\"

Option 2 — Project install

Scoped to one project only.

git clone https://github.com/Cybercentry/cybercentry-verification.git
mkdir -p .claude/skills
cp -r cybercentry-verification/skills/cybercentry-verification .claude/skills/

Use

In a Claude Code session, either:

• Invoke directly: /cybercentry-verification
• Or ask in natural language: Cybercentry, verify this codebase / run a security verification on this PR

Claude will load the skill, confirm scope and threat model, and produce a risk-prioritized report.

What's covered

OWASP Top 10:2025

• A01 Broken Access Control · A02 Security Misconfiguration · A03 Software Supply Chain Failures · A04 Cryptographic Failures · A05 Injection · A06 Insecure Design · A07 Authentication Failures · A08 Software or Data Integrity Failures · A09 Security Logging and Alerting Failures · A10 Mishandling of Exceptional Conditions

Plus:

• OWASP ASVS 5.0
• OWASP API Security Top 10
• OWASP Top 10 for LLM Applications 2025 (Gen AI Security Project)
• OWASP Cheat Sheet Series, WSTG, Code Review Guide

Contributing

PRs welcome. If you spot stale OWASP wording, missing categories, or have improvements to the verification flow, open an issue or PR.

License

MIT — fork freely.

---

Built by Cybercentry.