Skip to content
Back to skills

Security Review

ASecurity

Security-focused review rules for command and secret-sensitive changes.

  • 304 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added May 27, 2026
data-aishelldockergitsecurity

Security analysis

A100/100

Scanned May 27, 2026

npx -y skills add thewaltero/mythos-router --skill security-review --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Security Review?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Security Review
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/thewaltero-security-review/badge)](https://www.skillsdirectory.com/skills/thewaltero-security-review)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: security-review
version: 0.1.0
description: Security-focused review rules for command and secret-sensitive changes.
priority: 90
budget-multiplier: 1.0
allow-fallback: true
incompatible-with:
  - fast-mode
---

# security-review Skill

## Purpose
Use this skill when a task touches command execution, package scripts, CI, deploy files, authentication, secrets, receipts, memory, or user-controlled paths.

## Read First

If present, review these before making security-sensitive edits:

- SECURITY.md
- package.json
- .github/workflows/
- src/security-policy.ts
- src/commands/verify.ts
- any file that defines command execution, receipts, secrets, auth, or writable paths

## Rules
- Treat install scripts, lifecycle hooks, shell commands, workflows, Docker files, and deploy files as high-risk surfaces.
- Never expose secret values in logs, receipts, docs, tests, or examples.
- Prefer allowlists and structured parsing over broad string matching for security decisions.
- Do not weaken SWD verification, receipt integrity, dry-run behavior, or command review prompts.
- When a change introduces a new writable path, explain how it is constrained.

## Verification
- Check that sensitive outputs are redacted before storage.
- Check that command-affecting changes still require explicit human confirmation.
- Prefer read-only verification for CI-facing checks.

## Output

When finished, summarize:

- what security-sensitive files changed
- what risks were checked
- what verification was run
- any remaining assumptions or unverified areas

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.