Security Review

SkillSecurity

This skill gives your AI a set of fixed rules for spotting common security problems in code. Once added, it can review your code for missing authorization checks, unvalidated input, and ways that privileges could be escalated. You get a consistent security review without having to remember what to look for.

Available today. Use it from your connected AI after setup.

After adding it, point your AI at the code you want checked, such as a file or a set of recent changes, and ask it to run a security review. It will go through the code using the built-in rules and report what it finds.

Then ask your AI: use the Security Review skill

What your AI can do with it

  • Check code for missing authorization checks
  • Flag input that is not properly validated
  • Spot privilege escalation risks
  • Review changes against a fixed set of security rules
  • Apply the same checks every time code is reviewed

What this skill tells your AI

The instructions your AI receives, as published by bottelet/daybydaycrm in .claude/skills/security-review-checklist/SKILL.md and read by ahel’s review.

Purpose

Detect security risks in code during review phase. This skill does NOT enforce security. It identifies issues.


Scope

This skill evaluates:

  • authorization checks (missing or bypassed)
  • policy usage correctness
  • privilege escalation risks
  • unsafe controller or action exposure
  • validation gaps on external input

Ownership Boundary

Security Review does NOT:

  • implement policies
  • define roles/permissions
  • execute middleware logic
  • enforce runtime access control

Those belong to application security layers (Policies, Middleware, Gates).


Rules

  • Every sensitive action MUST have explicit authorization check
  • No unguarded resource actions (create/update/delete/view)
  • No direct access to privileged operations without policy validation
  • Input from external sources MUST be validated before use

Escalation Principle

If a potential security issue is detected:

  • assume it is a defect until proven otherwise
  • prioritize security over architectural convenience

Signals

GitHub stars
2k
Forks
767
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
security-review-checklist
Source
github.com/bottelet/daybydaycrm