code-review_skill

This skill performs a read-only code review to identify quality, style, and maintainability issues before security review.
  • TypeScript

0

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill matthew-plusprogramming/monorepo --skill code-review

  • SKILL.md10.9 KB

Overview

This skill reviews TypeScript implementations for code quality, style consistency, and best-practice adherence before security review. It is read-only: it reports findings with severity levels and recommendations but does not modify code. The output is a pass/block verdict plus an actionable report tied to atomic spec IDs.

How this skill works

The reviewer loads the spec group manifest and Implementation Evidence to determine the exact files changed, then inspects those files against a checklist covering style, maintainability, TypeScript best practices, error handling, API design, spec conformance, and tests. Findings are classified as Critical, High, Medium, or Low and compiled into a report with per-atomic-spec details and a recommended next action. If blocking issues exist, the skill marks the spec group blocked until fixes are applied and re-reviewed.

When to use it

  • After implementation finishes and /unify passes convergence
  • Mandatory for multi-file changes or public API additions
  • Required for changes to core services or any spec workflow completion
  • Optional for single-file bug fixes, docs-only, or test-only changes
  • Always run before the security review stage

Best practices

  • Only review files listed in atomic spec Implementation Evidence or the feature branch diff
  • Reference atomic spec IDs in every finding for traceability
  • Be specific: include file path, line range, severity, impact, and a suggested fix
  • Classify issues accurately so Critical/High block merge; Medium/Low are actionable recommendations
  • Report pre-existing issues as pre-existing and scope findings to changed files only

Example use cases

  • Run review on a spec group that implements a new public API to catch breaking changes and missing return types
  • Inspect multi-file auth flow changes to catch swallowed errors or improper localStorage use
  • Validate TypeScript typing and generic usage across services before security review
  • Assess test coverage and test quality for functions added by an atomic spec
  • Review UI component changes to ensure naming conventions and AC traceability are present

FAQ

PASS means no Critical/High issues were found; Medium/Low findings may remain and should be addressed in follow-ups.

What happens when High or Critical issues are found?

The skill marks the spec group BLOCKED and lists required fixes. Implementers must fix issues and re-run the review before security review.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational