Open-source AI agent skill for web security: reviews with proof, hardening playbooks per site type, black-box testing and privacy/cookie consent checks. Claude Code, Codex, Gemini CLI.
# Add to your Claude Code skills
git clone https://github.com/mayanthacreates/web-security-skillGuides for using ai agents skills like web-security-skill.
See how web-security-skill compares with popular alternatives.
web-security-skill is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by mayanthacreates. Open-source AI agent skill for web security: reviews with proof, hardening playbooks per site type, black-box testing and privacy/cookie consent checks. Claude Code, Codex, Gemini CLI. It has 0 GitHub stars.
web-security-skill's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/mayanthacreates/web-security-skill" and add it to your Claude Code skills directory (see the Installation section above).
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh web-security-skill against similar tools.
No comments yet. Be the first to share your thoughts!
Available after the catalog security scan passes.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM lists this project for discovery and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
An open-source AI agent skill for web security. It teaches your AI assistant to review, harden and black-box test websites the way a practitioner does: find the weakest reachable path, prove every finding, fail closed and build the cheapest wall first. It also covers privacy and cookie consent compliance for EU, UK and US visitors.
Install in one line (Claude Code):
/plugin install web-security --marketplace mayanthacreates/web-security-skill
skills/web-security/
├── SKILL.md philosophy, review table, decision framework,
│ universal baseline, router, black-box method,
│ incident response, review checklist
└── references/
├── static-marketing.md
├── wordpress-cms.md
├── ecommerce.md
├── saas-multitenant.md
├── marketplace-ugc.md
├── fintech-payments.md
├── healthcare-booking.md
├── apis-mobile-backends.md
└── privacy-consent-compliance.md
The core stays small. Your agent loads a playbook only when the site type matches, so a WordPress audit never loads the fintech rules.
Claude Code (plugin, recommended - gets updates):
/plugin install web-security --marketplace mayanthacreates/web-security-skill
or in two steps: /plugin marketplace add mayanthacreates/web-security-skill, then /plugin install web-security@mayanthacreates-security.
OpenAI Codex (plugin):
codex plugin marketplace add mayanthacreates/web-security-skill
codex plugin add web-security@mayanthacreates-security
Any agent with the open-source skills CLI:
npx skills add mayanthacreates/web-security-skill
Manual install:
git clone https://github.com/mayanthacreates/web-security-skill.git
cp -r web-security-skill/skills/web-security ~/.claude/skills/
For Codex use ~/.agents/skills/, for Gemini CLI use ~/.gemini/skills/. Restart the tool. The skill triggers on its own when a task matches, or you can ask for it by name.
Only test sites you own or have written permission to test. The black-box method is non-destructive by design: no exploitation, no load testing, no brute force beyond a small, controlled rate-limit check.
Issues and pull requests are welcome. Add facts only with a source and an "as of" date, and mark anything you couldn't confirm as unverified. Never include secrets or anything that identifies a real site you assessed.
MIT. See LICENSE. Also by the same author: marketing-frameworks-skill.