Skip to content
Back to search

AI_Project_Security

A modular suite of AI Agent Skills (SKILL.md) with 126 verifiable OWASP-aligned controls for application security reviews, threat modeling, and pre-release gates.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/kingsleyhk0323--ai-project-security/badge.svg)](/listings/kingsleyhk0323--ai-project-security)

HTML

<a href="/listings/kingsleyhk0323--ai-project-security"><img src="/api/listings/kingsleyhk0323--ai-project-security/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Clone the repository to obtain all four skills:

git clone [https://github.com/KingsleyHK0323/AI_Project_Security.git](https://github.com/KingsleyHK0323/AI_Project_Security.git)
cd AI_Project_Security

Install one or all skills into your AI assistant environment using each skill's canonical `name`:

<details> <summary><b>Claude Code (CLI)</b></summary>

Copy each `SKILL.md` into a directory matching its YAML `name` field inside your personal (`~/.claude/skills/`) or project (`.claude/skills/`) skills folder:

# Install all four skills globally
mkdir -p ~/.claude/skills/backend-security-checklist \
         ~/.claude/skills/backend-app-security-checklist \
         ~/.claude/skills/frontend-security-checklist \
         ~/.claude/skills/frontend-app-security-checklist

cp Backend_Security_Checklist/SKILL.md ~/.claude/skills/backend-security-checklist/SKILL.md
cp Backend_App_Security_Checklist/SKILL.md ~/.claude/skills/backend-app-security-checklist/SKILL.md
cp Frontend_Security_Checklist/SKILL.md ~/.claude/skills/frontend-security-checklist/SKILL.md
cp Frontend_App_Security_Checklist/SKILL.md ~/.claude/skills/frontend-app-security-checklist/SKILL.md

</details>

<details> <summary><b>Claude (Web &amp; Desktop Apps)</b></summary>

Package each skill folder (renamed to match its YAML `name`, containing `SKILL.md`) into a `.zip` file and upload it via **Settings → Capabilities / Skills**. Refer to Anthropic's documentation for current UI steps.

</details>

<details> <summary><b>Cursor, Gemini, ChatGPT, and Custom Agents</b></summary>

For tools that do not natively load `SKILL.md` frontmatter, copy the Markdown body of the relevant `SKILL.md` file(s) (everything below the closing `---` YAML delimiter) into your project's rules file (e.g., `.cursor/rules`), Gems/Custom Instructions, or system prompt.

</details>

<hr />

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 193 lines
  1. <div align="center">
  2. <h1>🛡️ AI Project Security</h1>
  3. <p>
  4. <b>A modular suite of production-grade AI Agent Skills for application security reviews, threat modeling, and pre-release security gates across backend, frontend, mobile, and desktop systems.</b>
  5. </p>
  6. <p>
  7. <img src="https://img.shields.io/github/license/KingsleyHK0323/AI_Project_Security?style=for-the-badge" alt="License: MIT" />
  8. <img src="https://img.shields.io/badge/Format-SKILL.md-blue?style=for-the-badge" alt="Format: SKILL.md" />
  9. <img src="https://img.shields.io/badge/Version-1.0.1-green?style=for-the-badge" alt="Version: 1.0.1" />
  10. <img src="https://img.shields.io/badge/Controls-126_Verifiable-black?style=for-the-badge" alt="Controls: 126 Verifiable" />
  11. </p>
  12. <p>
  13. <a href="#overview">Overview</a> •
  14. <a href="#skill-suite">Skill Suite</a> •
  15. <a href="#key-features">Key Features</a> •
  16. <a href="#installation">Installation</a> •
  17. <a href="#usage">Usage</a> •
  18. <a href="#repository-structure">Structure</a> •
  19. <a href="#contributing">Contributing</a> •
  20. <a href="#license">License</a>
  21. </p>
  22. </div>
  23. <hr />
  24. ## Overview
  25. **AI Project Security** is an open-source suite of four interoperable `SKILL.md` specifications designed to equip AI coding assistants with structured, defense-in-depth application security checklists. Spanning **40 security domains** and **126 verifiable controls** aligned with OWASP Top 10, OWASP API Security Top 10, OWASP ASVS, and OWASP MASVS, the suite supports secure feature design, evidence-based code and architecture reviews, and pre-release security gates.
  26. > [!IMPORTANT]
  27. > **Defensive Scope Only:** These skills are built strictly for auditing, hardening, and reviewing systems you own or are authorized to assess. They enforce evidence-based reporting (`Pass`, `Partial`, `Fail`, `N/A`, `Not verified`) and prohibit generating working exploits or attack payloads.
  28. <hr />
  29. ## Skill Suite
  30. Select and install the skills that match your application architecture. Companion skills cross-reference each other without duplicating findings:
  31. | Skill Directory | Skill Name (`SKILL.md`) | Target Architecture | Coverage |
  32. |---|---|---|---|
  33. | [`Backend_Security_Checklist/`](Backend_Security_Checklist/SKILL.md) | `backend-security-checklist` | General-purpose server-side services, REST/GraphQL/RPC APIs, background workers, databases, and CI/CD pipelines | 10 domains • 35 controls |
  34. | [`Backend_App_Security_Checklist/`](Backend_App_Security_Checklist/SKILL.md) | `backend-app-security-checklist` | Backend APIs powering native mobile and consumer apps (device binding, anti-replay signing, push notifications, deep links, forced upgrades) | 10 domains • 28 controls |
  35. | [`Frontend_Security_Checklist/`](Frontend_Security_Checklist/SKILL.md) | `frontend-security-checklist` | Browser-based web applications, SPAs, SSR/SSG apps, and PWAs (React, Vue, Angular, Next.js, Nuxt, Svelte, vanilla JS) | 10 domains • 37 controls |
  36. | [`Frontend_App_Security_Checklist/`](Frontend_App_Security_Checklist/SKILL.md) | `frontend-app-security-checklist` | Native mobile (iOS, Android), cross-platform (React Native, Flutter), WebView, and desktop (Electron, Tauri) clients | 10 domains • 26 controls |
  37. <hr />
  38. ## Key Features
  39. <table>
  40. <tr>
  41. <td width="50%">
  42. <h3>🎯 Three Operational Modes</h3>
  43. <p>Adapts seamlessly to <b>Design Guidance</b> (controls per feature), <b>Code &amp; Architecture Review</b> (severity-rated findings table), and <b>Release Gates</b> (domain-grouped task checklists).</p>
  44. </td>
  45. <td width="50%">
  46. <h3>📊 Evidence-Based Auditing</h3>
  47. <p>Evaluates every applicable control with explicit status tags (<code>Pass</code>, <code>Partial</code>, <code>Fail</code>, <code>N/A</code>, <code>Not verified</code>) and severity ratings (<code>Critical</code> to <code>Info</code>).</p>
  48. </td>
  49. </tr>
  50. <tr>
  51. <td width="50%">
  52. <h3>🔗 Full-Stack Cross-Referencing</h3>
  53. <p>Links client-side defenses (such as route guards, CSRF tokens, and deep links) directly to their required server-side enforcement counterparts.</p>
  54. </td>
  55. <td width="50%">
  56. <h3>🧭 Industry-Standard Alignment</h3>
  57. <p>Maps all 126 traceable control IDs (e.g., <code>AUTHZ-01</code>, <code>INJ-01</code>, <code>XSS-02</code>, <code>STORE-01</code>) to OWASP ASVS, MASVS, Top 10, and API Security Top 10.</p>
  58. </td>
  59. </tr>
  60. </table>
  61. <hr />
  62. ## Installation
  63. Clone the repository to obtain all four skills:
  64. ```bash
  65. git clone [https://github.com/KingsleyHK0323/AI_Project_Security.git](https://github.com/KingsleyHK0323/AI_Project_Security.git)
  66. cd AI_Project_Security
  67. ```
  68. Install one or all skills into your AI assistant environment using each skill's canonical `name`:
  69. <details>
  70. <summary><b>Claude Code (CLI)</b></summary>
  71. Copy each `SKILL.md` into a directory matching its YAML `name` field inside your personal (`~/.claude/skills/`) or project (`.claude/skills/`) skills folder:
  72. ```bash
  73. # Install all four skills globally
  74. mkdir -p ~/.claude/skills/backend-security-checklist \
  75. ~/.claude/skills/backend-app-security-checklist \
  76. ~/.claude/skills/frontend-security-checklist \
  77. ~/.claude/skills/frontend-app-security-checklist
  78. cp Backend_Security_Checklist/SKILL.md ~/.claude/skills/backend-security-checklist/SKILL.md
  79. cp Backend_App_Security_Checklist/SKILL.md ~/.claude/skills/backend-app-security-checklist/SKILL.md
  80. cp Frontend_Security_Checklist/SKILL.md ~/.claude/skills/frontend-security-checklist/SKILL.md
  81. cp Frontend_App_Security_Checklist/SKILL.md ~/.claude/skills/frontend-app-security-checklist/SKILL.md
  82. ```
  83. </details>
  84. <details>
  85. <summary><b>Claude (Web &amp; Desktop Apps)</b></summary>
  86. Package each skill folder (renamed to match its YAML `name`, containing `SKILL.md`) into a `.zip` file and upload it via **Settings → Capabilities / Skills**. Refer to Anthropic's documentation for current UI steps.
  87. </details>
  88. <details>
  89. <summary><b>Cursor, Gemini, ChatGPT, and Custom Agents</b></summary>
  90. For tools that do not natively load `SKILL.md` frontmatter, copy the Markdown body of the relevant `SKILL.md` file(s) (everything below the closing `---` YAML delimiter) into your project's rules file (e.g., `.cursor/rules`), Gems/Custom Instructions, or system prompt.
  91. </details>
  92. <hr />
  93. ## Usage
  94. Once installed, prompt your assistant naturally when writing, reviewing, or shipping code:
  95. ```text
  96. Review this FastAPI order endpoint for security issues, focusing on object-level
  97. authorization, mass assignment, and error handling.
  98. ```
  99. ```text
  100. Generate a pre-release security checklist for our React + Next.js web app
  101. covering CSP headers, token storage, and third-party scripts.
  102. ```
  103. ```text
  104. Audit our iOS Swift authentication flow and Universal Links configuration
  105. before App Store submission.
  106. ```
  107. <details>
  108. <summary><b>Standard Security Review Output Structure</b></summary>
  109. When performing a code or architecture review, each skill produces a structured report containing:
  110. 1. **Summary:** Scope, overall security posture, finding counts by severity, and top risks.
  111. 2. **Scope and Assumptions:** Reviewed files/configurations, unverified surfaces, and stack assumptions.
  112. 3. **Findings Table:** Traceable Control ID, Control Name, Severity (`Critical`, `High`, `Medium`, `Low`, `Info`), Status, Evidence, and Recommendation.
  113. 4. **Detailed Remediation:** Concrete minimal code or configuration fixes with verification steps.
  114. 5. **Passed Controls & Next Steps:** Verified controls and prioritized action items.
  115. </details>
  116. <hr />
  117. ## Repository Structure
  118. ```text
  119. .
  120. ├── Backend_App_Security_Checklist/
  121. │ └── SKILL.md # Backend controls for mobile & consumer apps (10 domains, 28 controls)
  122. ├── Backend_Security_Checklist/
  123. │ └── SKILL.md # General-purpose backend & API security baseline (10 domains, 35 controls)
  124. ├── Frontend_App_Security_Checklist/
  125. │ └── SKILL.md # Native mobile, hybrid, WebView & desktop app controls (10 domains, 26 controls)
  126. ├── Frontend_Security_Checklist/
  127. │ └── SKILL.md # Browser-based web frontend & PWA security controls (10 domains, 37 controls)
  128. ├── .github/ # Repository workflows and funding configuration
  129. ├── README.md # Suite documentation, comparison matrix, and installation guide
  130. └── LICENSE # MIT License
  131. ```
  132. <hr />
  133. ## Contributing
  134. Contributions are welcome, especially updates reflecting new platform security guidance (such as Android Keystore/Credential Manager updates, browser CSP/Trusted Types changes, or OWASP standard revisions).
  135. 1. Fork the repository and create a feature branch (`git checkout -b docs/update-owasp-mapping`).
  136. 2. Edit the relevant `SKILL.md` (and `README.md` if control counts or structure change), keeping YAML frontmatter valid.
  137. 3. Ensure all new controls include a unique ID, clear requirement, verification steps, and defensive-only scope.
  138. 4. Open a pull request describing what changed and why.
  139. <hr />
  140. ## License
  141. Released under the [MIT License](LICENSE). Copyright (c) 2026 Kingsley ([@KingsleyHK0323](https://github.com/KingsleyHK0323)).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.