ClaudeSuperPower
SubmitLog in

secure-by-design

Skill

Run an enterprise security review of a system design or existing code before it ships. Use when the user says "security review", "secure by design", "threat model this", "is this design secure", "what security controls do we need", "audit this for security", "harden this system", "RBAC design", "rev

0Code Review

Install

git clone https://github.com/ooiyeefei/ccc.git ~/.claude/skills/secure-by-design

What is secure-by-design?

Run an enterprise security review of a system design or existing code before it ships. Use when the user says "security review", "secure by design", "threat model this", "is this design secure", "what security controls do we need", "audit this for security", "harden this system", "RBAC design", "review my auth", or "enterprise security checklist". Classifies assets and trust boundaries FIRST, then recommends only the controls that fit — grounded in battle-tested patterns. Right-sizes for team scale (bones now, ceremony later). Do NOT use for writing a specific exploit, penetration-testing a system the user does not own, or general non-security code review.

What this can do

Capabilities declared in this component's own frontmatter — not inferred.

Inherit all session tools

Declares no tool restrictions — inherits every session tool

~166 tokens of context used while enabled, before you invoke anything

Documentation

README · ~4 min read

Secure by design

A review partner for the security of any system — an RBAC model, a multi-tenant SaaS, an API, an agent. You classify what you are protecting before proposing controls, then recommend only what fits the assets, trust boundaries, and team scale. Output: a short, ranked list of applicable controls with the reasoning, not a generic checklist dump.

The leading words below are load-bearing. Use them in your reasoning — they carry the decisions.

Leading words (say these)

  • deny-by-default — an unmapped permission is denied, never allowed.
  • fail-closed — unknown state ⇒ deny; an error path never grants access.
  • least privilege — the smallest capability that does the job; unknown/absent role ⇒ the least-privileged role.
  • two axes (scope × role) — which resources vs what actions are independent; do not collapse siblings onto one ladder.
  • anti-IDOR — never trust a client-supplied id as a grant; resolve the caller's own scope independently.
  • append-only audit — one immutable event per privileged action, in the same transaction, actor bound from auth (never a client arg).

Reviews

Log in to leave a review.

No reviews yet — be the first.

Explore related

Other things in this space — across every part of the ecosystem, not just skills.

Subagents

All subagents →

ECC

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

235.3K stars

code-reviewer

Use this agent when you need to review code for adherence to project guidelines, style guides, and best practices. This agent should be used proactively after writing or modifying code, especially before committing changes or creating pull requests. It will check for style violations, potential issu

32.8K stars

agent-creator

Use this agent when the user asks to "create an agent", "generate an agent", "build a new agent", "make me an agent that...", or describes agent functionality they need. Trigger when user wants to create autonomous agents for plugins. Examples: <example> Context: User wants to create a code review a

32.8K stars