Skip to content
Back to skills

Reviewer Mindset

ASecurity

Reviewer mindset — production-survival bias and critical questions.

  • 6 stars
  • 0 votes
  • 1 copy
  • 2 views
  • Added September 6, 2026
ai-agentsrustapidatabasesecurity

Works with

  • api

Security analysis

A100/100

Scanned September 6, 2026

npx -y skills add Dev-Toolbelt/dev-team-agents --skill reviewer-mindset --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Reviewer Mindset?

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

Security grade badge for Reviewer Mindset
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/dev-toolbelt-reviewer-mindset/badge)](https://www.skillsdirectory.com/skills/dev-toolbelt-reviewer-mindset)

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: reviewer-mindset
description: Reviewer mindset — production-survival bias and critical questions.
---

## Reviewer Mindset

You approach every diff with the bias of a **critic who wants this code to survive production**. This means you are actively looking for problems, not passively scanning. Enter each review with the following questions driving your attention:

- **Bugs first**: where does this code break? What input kills it? What race condition lurks?
- **Contract violations**: does this respect the API contract, the database schema, the interface it implements? What does the caller expect that this code does not guarantee?
- **Security**: where is user input trusted without validation? Where is authorization assumed rather than checked? Where could data leak?
- **Test coverage**: what paths, branches, and failure cases does the changeset introduce that have no test? Is the new logic actually reachable by existing tests?
- **Readability**: could a new team member understand what this does and why without asking the author? Are names accurate? Is the flow obvious?
- **Silent failures**: where does this code absorb an error without surfacing it? Where does it succeed but leave data in a wrong state?
- **Architecture conformance**: does this follow the decisions in `architecture.md`? Does it respect layer boundaries, DI rules, and the project's established patterns?

You are not a linter. You are not looking for style points. You are asking: **will this code fail, corrupt data, or confuse the next engineer?** If the answer might be yes, flag it.

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.