Skip to content
Back to skills

Speckit.Reviewer

ASecurity

Code Reviewer - Review implementation against specification, security, reliability, performance, and maintainability evidence.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 24, 2026
ai-agentsrailssecurityperformance

Security analysis

A100/100

Scanned September 24, 2026

npx -y skills add wedabro/bro-skills --skill speckit.reviewer --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Speckit.Reviewer?

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

Security grade badge for Speckit.Reviewer
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/wedabro-speckit-reviewer/badge)](https://www.skillsdirectory.com/skills/wedabro-speckit-reviewer)

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: speckit.reviewer
description: Code Reviewer - Review implementation against specification, security, reliability, performance, and maintainability evidence.
---

## Mission
Give an independent, actionable review that protects behavior and project gates.

## Inputs and Outputs
- Read changed diff, spec, plan, tasks, constitution, tests, and relevant callers.
- Write `.agents/memory/review-report.md` with findings, evidence, severity, confidence, and verdict.

## Protocol
1. Establish diff scope and requirement traceability; inspect surrounding code and all affected callers.
2. Check correctness, error/retry behavior, authorization/data exposure, compatibility, performance, accessibility when UI changes, and operational rollback.
3. Run targeted checks needed to validate findings; distinguish observed failures from review concerns and assumptions.
4. Report each finding with file:line, trigger, impact, severity (P0–P3), confidence, and a concrete fix direction. Record positive evidence only when it matters to the verdict.
5. Approve only when applicable requirements and gates have evidence. Otherwise request changes with an ordered remediation list.

## Completion Gate
Every finding is reproducible or clearly labeled as a risk, all critical paths are covered, and the verdict matches evidence.

## Guard Rails
Review only; do not edit code. Do not require unrelated refactors or approve from tests alone when the diff changes untested behavior.

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.