Cyber Neo skill
A Claude Code skill that audits a project against OWASP 2025 and CWE Top 25 and writes a prioritized report with fixes.
- 297 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Reads the whole project, including discovered secrets, and saves a report with code excerpts to the Desktop.
- Runs installed Semgrep, Trivy, Gitleaks and package manager audits.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/cyber-neoDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Reads the whole project, including discovered secrets, and saves a report with code excerpts to the Desktop.
- Runs installed Semgrep, Trivy, Gitleaks and package manager audits.
This is third-party code. Review the repository files before installing.
What it does
Cyber Neo is a Claude Code skill that runs a security audit of a local project. The command `/cyber-neo /path/to/your/project` covers 11 areas: code (SAST), authentication and authorization, cryptography, secrets, dependencies with known CVEs, web security, supply chain, CI/CD, Docker, error handling and logging. Five parallel subagents do the analysis. The report with an overall severity score, severity levels, CWE and OWASP references and fix examples is saved to `~/Desktop/cyber-neo-report-{project-name}-{YYYY-MM-DD}.md`. The skill reads the project and writes only that file.
Who it is for. Developers and small teams that want a quick security audit without a separate scanner or security expertise.
Good fit when
- Before a production release the project should be checked against OWASP 2025 and CWE Top 25.
- Keys for AWS, Stripe or GitHub may have been left in the repository, and 60-plus patterns look for them.
- Dependencies and lock files are checked for known CVEs and package substitution.
- GitHub Actions workflows and Dockerfiles are reviewed for excess permissions and secrets in layers.
Not a fit when
- You need to test someone else's or a running system, because the skill reads local code only.
- A formal certified audit or a pentest for a regulator is required.
- The project is in Go, Ruby, Java, Rust or PHP, since language references exist only for JavaScript, TypeScript and Python for now.
Example request
Audit this project with `/cyber-neo .` and name the three most urgent fixes.Limitations
Language references exist for JavaScript, TypeScript and Python. The README defers Go, Ruby, Java, Rust and PHP to version 0.2, so generic patterns apply meanwhile. Semgrep, Trivy, Gitleaks, pip-audit and cargo-audit are optional, and without them Claude alone runs the check at lower depth. In projects of 10,000 files or more the skill reviews critical paths (entry points, authorization, configuration, dependencies) and states the coverage share in the report. Findings come from pattern review and a model, so false positives and misses are possible. Installation through `/plugin install cyber-neo` is marked as coming soon in the README.
How to disable. Delete the `~/.claude/skills/cyber-neo` directory or the symlink to it. Remove finished reports from the Desktop separately.
FAQ
What does the skill change in my project?
Nothing. The README sets a read-only rule: files stay untouched, no code or package installs run, and a single report file is written to the Desktop.
Do I need Semgrep and Trivy?
No. Without them the skill uses Claude analysis and its own Python secret scanner. It picks up installed Semgrep, Trivy and Gitleaks automatically for deeper results.
How do I start a scan?
Run `/cyber-neo /path/to/your/project` or `/cyber-neo .` in the working directory. If the skill is installed correctly, running it without a path makes it ask for the project path.
How does the skill behave on a large project?
It splits projects into three tiers by file count. Up to 1,000 files are all scanned, from 1,000 to 10,000 it prioritizes `src/`, `app/` and configuration, above 10,000 it reviews entry points, authorization and dependencies.
Which languages are covered?
Reference files exist for JavaScript, TypeScript and Python, and generic SAST patterns apply to other languages. The README promises Go, Ruby, Java, Rust and PHP in version 0.2.
Discussion
No questions or reviews yet.
Related
SkillsA Claude Code skill that finds vulnerabilities in one session: threat model, verified findings, CWE and CVSS report.
Security audit skill with GOST and FSTEC
Security Audit Skill for Claude Code
A security audit skill for Claude Code: OWASP Top 10 plus GOST R 56939, FSTEC Order No. 21 and 152-FZ, parallel subagents and a compliance matrix
Claude Code CyberSecurity Skills
Claude Code CyberSecurity Skill Collection
A set of 22 cybersecurity skills for Claude Code: recon, vulnerability assessment, incident response, red and blue team, malware analysis
Cybersecurity Skills
Cybersecurity Skills for AI Agents
42 security skills for Claude Code, Cursor and Codex: code and cloud audits, authorization-gated pentesting and incident response
A code security audit skill by Cloudflare: the agent runs recon, coverage-led hunting and independent verification of findings, then produces a structured repor
A code security plugin: finds vulnerabilities, challenges its findings, proves exploitability and fixes with a verified patch
README in short
The skill combines a 7-phase `SKILL.md`, 14 vulnerability reference files and two dependency-free Python scripts: a secret scanner and a lock-file check.