prisma-check

📁 htlin222/dotfiles 📅 5 days ago
9
总安装量
9
周安装量
#33505
全站排名
安装命令
npx skills add https://github.com/htlin222/dotfiles --skill prisma-check

Agent 安装分布

gemini-cli 8
opencode 8
codebuddy 8
github-copilot 8
codex 8
kimi-cli 8

Skill 文档

PRISMA 2020 Compliance Checker

Audit systematic review and meta-analysis manuscripts against the PRISMA 2020 (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 27-item checklist.

Workflow

  1. Read the full manuscript
  2. Confirm the study is a systematic review and/or meta-analysis
  3. Walk through each item below, including all sub-items
  4. For each item, assign: Reported / Partial / Missing / N/A
  5. Quote the relevant manuscript text as evidence
  6. Output a compliance summary + actionable fixes for Missing/Partial items
  7. Check for PRISMA flow diagram (strongly recommended)

PRISMA 2020 Checklist (27 Items)

Title and Abstract

# Topic Requirement
1 Title Identify the report as a systematic review
2 Abstract Structured summary: background, objectives, data sources, eligibility criteria, participants/interventions, appraisal/synthesis methods, results, limitations, conclusions, registration number

Introduction

# Topic Requirement
3 Rationale Describe rationale in context of existing knowledge
4 Objectives Explicit statement of objectives or questions the review addresses

Methods

# Topic Requirement
5 Eligibility criteria Inclusion/exclusion criteria and how studies were grouped for syntheses
6 Information sources All databases, registers, websites, organisations, reference lists searched; date of last search for each
7 Search strategy Full search strategies for all databases, including filters and limits
8 Selection process Methods to decide inclusion: number of reviewers per record, independence, automation tools
9 Data collection process Methods to collect data: number of reviewers, independence, processes for confirming with investigators, automation tools
10a Data items — outcomes List and define all outcomes sought; whether all compatible results per domain were sought
10b Data items — other variables List all other variables sought (participant/intervention characteristics, funding); assumptions about missing/unclear data
11 Risk of bias assessment Methods to assess risk of bias: tool(s) used, number of reviewers, independence, automation
12 Effect measures Effect measure(s) for each outcome (risk ratio, mean difference, etc.)
13a Synthesis — eligibility Processes to decide which studies eligible for each synthesis
13b Synthesis — data preparation Methods to prepare data for synthesis (handling missing statistics, conversions)
13c Synthesis — display Methods to tabulate or visually display results
13d Synthesis — statistical methods Methods to synthesise results with rationale; if meta-analysis: model(s), heterogeneity methods, software
13e Synthesis — heterogeneity Methods to explore causes of heterogeneity (subgroup analysis, meta-regression)
13f Synthesis — sensitivity Sensitivity analyses to assess robustness
14 Reporting bias assessment Methods to assess risk of bias from missing results (reporting biases)
15 Certainty assessment Methods to assess certainty in the body of evidence (e.g., GRADE)

Results

# Topic Requirement
16a Study selection Results of search and selection, ideally with a PRISMA flow diagram
16b Study selection Cite excluded studies that appeared to meet criteria; explain why excluded
17 Study characteristics Cite each included study and present its characteristics
18 Risk of bias in studies Present risk of bias assessments for each study
19 Individual study results For all outcomes: summary statistics per group, effect estimate with precision (CI), structured tables or forest plots
20a Synthesis results Briefly summarise characteristics and risk of bias among contributing studies
20b Synthesis results All statistical syntheses: summary estimates with CI, heterogeneity measures, direction of effect
20c Synthesis results Investigations of heterogeneity causes
20d Synthesis results Sensitivity analysis results
21 Reporting biases Risk of bias from missing results for each synthesis
22 Certainty of evidence Certainty assessments for each outcome

Discussion

# Topic Requirement
23a Discussion General interpretation in context of other evidence
23b Discussion Limitations of the evidence
23c Discussion Limitations of the review processes
23d Discussion Implications for practice, policy, and future research

Other Information

# Topic Requirement
24a Registration Registration name and number, or state not registered
24b Protocol Where protocol can be accessed, or state not prepared
24c Protocol amendments Describe and explain amendments to registration or protocol
25 Support Financial/non-financial support; role of funders/sponsors
26 Competing interests Declare competing interests of review authors
27 Data/code availability Report availability of: data collection forms, extracted data, analytic code, other materials

Critical PRISMA Elements

These are the items journals most commonly reject manuscripts for missing:

Must-Have Why
Flow diagram (16a) Most journals will not review without one
Full search strategy (7) Reproducibility requirement; include as appendix if long
PROSPERO registration (24a) Increasingly mandatory; register before data extraction
Risk of bias table (18) Required for credibility; use RoB 2 or NOS as appropriate
GRADE or certainty (15, 22) Expected by top-tier journals

Common PRISMA Gaps

Frequently Missing Fix
Item 7 (Full search strategy) Append complete search strings for each database
Item 8 (Selection process detail) State number of reviewers, kappa/agreement, screening tool
Item 13d (Heterogeneity methods) Specify I-squared thresholds, random/fixed-effects model choice
Item 14 (Reporting bias) Add funnel plot + Egger’s test or trim-and-fill
Item 16b (Excluded studies) List near-misses with exclusion reasons
Item 24a (Registration) Register on PROSPERO; if post-hoc, disclose it

Output Format

PRISMA 2020 Compliance Report
Manuscript: [filename]

Summary: X/27 Reported | Y Partial | Z Missing | W N/A

CRITICAL MISSING (journal will likely reject):
  [Item #] [Topic] — [What's needed]

OTHER MISSING:
  [Item #] [Topic] — [What's needed]

PARTIAL ITEMS:
  [Item #] [Topic] — [What's present] → [What's missing]

Flow diagram: [Present / Missing]
Registration: [Registered (ID) / Not registered / Not stated]

Related Skills

  • /manuscript — Overall manuscript writing and anti-pattern scanning
  • /meta-manuscript-assembly — Assemble tables, figures, references for meta-analyses