# Systematic Review

> Conduct systematic reviews with transparent search, screening, and synthesis protocols that minimize bias in evidence summaries. Use when high-stakes decisions require rigorous aggregation of existing studies (academic, clinical, policy, or technical).

- Skill: `itsual/systematic-review` (Agent Skill)
- Install (CLI): `npx skillmds@latest add itsual/systematic-review`
- Raw SKILL.md: https://api.skillmd.com/api/skills/itsual/systematic-review/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: itsual (https://skillmd.com/u/itsual)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/itsual/systematic-review

---


# Systematic Review

## Overview

A systematic review answers a defined question by finding, appraising, and synthesizing all relevant evidence using a predetermined protocol — reducing cherry-picking.

## When to Use

- High-stakes evidence summaries
- Academic and clinical evidence bases
- Policy and standards decisions
- Technical state-of-the-art with auditability needs

## Core Practices

- Register or document the protocol (question, criteria, search plan)
- Search multiple sources with recorded strategies
- Screen studies with dual review when possible
- Appraise quality/risk of bias
- Synthesize narratively or quantitatively (meta-analysis) as appropriate
- Report PRISMA-style flow and limitations

## Principles

- Protocol precedes results to limit bias
- Absence of evidence ≠ evidence of absence
- Quality of included studies bounds quality of conclusions
- Updates may be needed as literature moves

## Verification

- [ ] Protocol and criteria are explicit
- [ ] Search and screening are reproducible
- [ ] Conclusions respect study quality and heterogeneity