ai-saas-security-review

Community

Audit AI SaaS security across all threat layers

Authortlennon-ie
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill helps you reliably assess whether an AI SaaS is protected against common web security failures and AI-specific abuse (like prompt injection, jailbreaks, and system-prompt extraction), before vulnerabilities become incidents.

Core Features & Use Cases

  • Threat-layer security review: Organizes findings across identity, authorization, input, AI, action, and observability layers so severity judgments stay consistent.
  • OWASP + AI-specific checks: Covers classical web risks (authz, injection, XSS, CSRF, uploads, redirects, headers) plus AI risks (direct/indirect injection via memory or uploads, tool-output injection, jailbreak patterns).
  • Production hardening procedures: Provides concrete audit playbooks for RLS, webhook verification, secret scanning, rate limiting, admin surface protection, and a structured finding format.

Quick Start

Use this skill to review a pull request by mapping each code change to the six threat layers and producing CRITICAL/HIGH findings with repro steps and actionable fixes.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: ai-saas-security-review
Download link: https://github.com/tlennon-ie/product-building-agents/archive/main.zip#ai-saas-security-review

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.