编程

Meta Analysis / 医学Meta分析

Comprehensive R-based meta-analysis skill covering RevMan 5.x + Stata equivalents (metareg/mvmeta) + esc + RVE + Bayesian NMA (Stan/JAGS) + survival meta + TSA + single-group meta + diagnostic meta + systematic review workflow; produces forest plots, funnel plots, heterogeneity (I²), publication bias, subgroup analysis, meta-regression, network meta, for a total of 23 analysis figures. Auto-switches language (defaults to English, switches to Chinese in zh-* environments). All analyses ship reproducible R code. / 基于 R 的全方位 Meta 分析技能，覆盖 RevMan 全部功能 + Stata 等价（metareg/mvmeta）+ esc + RVE + 贝叶斯 NMA（Stan/JAGS）+ 生存 Meta + TSA + 单组率 Meta + 诊断 Meta + 系统评价流程；输出森林图、漏斗图、异质性(I²)、发表偏倚、亚组分析、元回归、网络 Meta等共 23 种分析图形。中英双语自动切换（默认英文/中文环境切中文），所有分析提供可复现 R 代码。

它能做什么

技能文档

Meta-Analysis / 医学Meta分析

R-based comprehensive meta-analysis. Every module ships reproducible R code.

Language / 语言

This skill is English-default; auto-switches to Chinese when OS locale is zh-*.
Detailed language policy → references/language_policy.md.
Chinese user guide → README_zh-CN.md
English user guide → README.md

Triage (ct-base §5.2)

On first user message, classify into one of three:

Classification	Condition	Action
Simple	Single, specific intent (e.g., "pool OR from these 5 studies")	Reply directly, no menu
Complex	Multi-decision / multi-parameter (e.g., "network meta with 3 interventions, subgroup, check inconsistency")	Present level-1 routing menu incl. "③ Can't decide? → explain the differences between these choices" entry; full menu → references/interactive_menu.md
Vague	Unclear what user wants (e.g., "I need meta-analysis help")	Grill-me style branch questions, 1–3 per round

If unsure between Simple and Complex → give short reply + optional expansion hint.

Traceability / Grounding (ct-base §5.1)

All factual/assertive claims must cite source: specific ref-*.md section (e.g., §3.6) or official guideline. If a claim has no verifiable source → mark ⚠️ official verify and ask user to confirm against official text.

Initialization

Integrity check: R source code is embedded as Python templates (scripts/r_*.py + r_templates.py) to comply with marketplace restrictions on .R files. On first use, run:
```
bash scripts/check_integrity.sh   # exit 0 = ready
```
- If scripts/*.R exists → ready; if missing → script auto-runs python r_templates.py to generate from templates.
R env: Rscript --version → missing → https://cran.r-project.org/
Install (manual): This skill does NOT auto-install R packages. Run bash scripts/check_integrity.sh or pre-analysis self-check to see missing packages; then manually install in R, or Rscript scripts/setup_packages.R to view full list.
- Packages: metafor meta dmetar netmeta ggplot2 esc clubSandwich robumeta bayesmeta multinma gemtc dosresmeta survmeta mada metagear gridExtra ggforestplot svglite
- ⚠️ Manual install connects to CRAN; use trusted network.
Workspace: Create meta_analysis/ + output/ in current directory (⚠️ will write files).
Memory: read ~/.workbuddy/MEMORY.md for R config (only R-related config keys; unrelated personal content is ignored and never sent anywhere).

Interactive Guide

Triage path: Vague → Level 1 menu (7 categories). Select → Level 2 with data-format hints. Sufficient info → skip menu, run analysis directly.

=== Level 1: Main ===
1️⃣ Pairwise Meta     4️⃣ Effect Size      7️⃣ Review Workflow
2️⃣ Heterogeneity & Bias  5️⃣ Visualization
3️⃣ Advanced Models   6️⃣ Study Quality

=== Level 2: Sub-Menus (excerpt) ===
[1] Pairwise: Binary(OR/RR/RD) | Continuous(SMD/MD) | Pre-calc(yi+CI) | Survival(HR/IRR) | Correlation(r→Zr)
[2] Heterogeneity: I²/Q/τ² | Subgroup | Meta-regression | Pub Bias(Egger/Begg/Trim-fill) | Sensitivity | GOSH | Baujat
[3] Advanced: Multi-arm NMA(netmeta) | Bayesian NMA Stan(multinma) | JAGS(gemtc) | Multilevel | Multivariate | IPD | Dose-Resp(dosresmeta) | Survival(survmeta) | TSA | Bootstrap
[4] Effect Size: Mean→d | t/F/r→d | d↔Hedges'g | d↔logOR | r↔Z | OR↔logOR | Batch
[5] Viz: Forest(5 themes) | Funnel | Bubble | GOSH | Network | League Table | RoB Traffic-light | Power Curve | Drapery
[6] Quality: RoB 1.0/2.0 | ROBINS-I | GRADE | PRISMA Checklist
[7] Workflow: PRISMA Flow | Screening GUI | PDF Batch-download (⚠️ opt-in) | Digitize | NNT Meta

Data formats & full details → references/interactive_menu.md

Other formats? Install @skill:statdata-transfer for 50+ format conversion.

Core Functions

Module	R Packages & Functions
Single-Group Meta	`metaprop()` `metamean()` `metainc()` `metacor()` `metarate()`
Pairwise Meta	`metabin()` `metacont()` `metagen()` `rma()` — FE/RE(DL/HK)/MH/Peto
Effect Size	`escalc()` `esc_mean_sd()` — SMD, OR, RR, RD, HR, ROM, ZCOR
Forest/Funnel	`forest()` `funnel()` + ggplot2 (5 themes)
Heterogeneity	I², Q, τ², H², 95% PI — auto-reported
Publication Bias	`regtest()` `ranktest()` `trimfill()` `selmodel()`
Subgroup & Reg	`rma(mods=~factor-1)` + `bubble()`
Sensitivity	Leave-one-out, Cumul, GOSH, quality filter
Bayesian Pairwise	`bayesmeta::bayesmeta()` — half-normal/JC prior
Bayesian NMA	`run_bayes_nma_multinma()`(Stan) · `run_bayes_nma_gemtc()`(JAGS)
Multilevel/MV Meta	`rma.mv()` + UN/CS/HCS/AR1/ID/DIAG
Survival Meta	`run_surv_meta()` — survmeta wrapper
TSA & Diagnostics	`run_tsa()` · `baujat()` `drapery()` `bootmeta()`
Dose-Response	`run_dose_resp()` — dosresmeta wrapper
Diagnosis Meta	`mada::reitsma()` bivariate + SROC
RVE Robust	`robumeta::robu()` `clubSandwich::vcovCR(CR2)`
Review Workflow	`metagear`: PRISMA, screen, PDF, digitize
Quality	`rob()` RoB 1.0/2.0/ROBINS-I + GRADE
Power	`run_power_curve()` + subgroup power

Reusable API (Mandatory)

Any analysis MUST call existing functions — never rewrite the full pipeline inline. Full function list + calling examples → references/advanced_api.md.

Security & Scope

Execution model: R analysis runs locally; skill does NOT auto-install R packages (missing → list + prompt manual install). PDF full-text download from external services ONLY on explicit user instruction. Analysis artifacts written to meta_analysis/ + output/ by default.

Not clinical judgment: Results require professional interpretation.

No literature DB search: Does not search literature databases; only downloads full text when user provides DOI/PMID.

Output

analysis_complete.R + forest/funnel (.svg+.png) + results_summary.md + data_backup.csv.

Output figures as editable SVG; editing + journal format conversion → references/svg_editing.md.

Units

See references/units.md for the full 5-field schema (input / output / dependency / AI autonomy / consumer units).

References

File	Content
`interactive_menu.md`	How to use in a chat (user-friendly guide, 5 conversation examples)
`ADVANCED.md`	Advanced reference (CLI, formulas, troubleshooting, full file structure)
`ADVANCED_zh-CN.md`	Advanced reference (Chinese version)
`revman_complete.md`	RevMan→R 1:1 code mapping
`stata_to_r_mapping.md`	Stata metareg/mvmeta→R equivalents
`esc_robust_meta.md`	Effect size conversions + RVE reference
`advanced_analysis.md`	Multilevel/IPD/Bayesian/Dose-Resp/Power
`single_group_meta.md`	metaprop/metamean/metainc/metacor + NNT
`bayesian_nma.md`	multinma (Stan) + gemtc (JAGS) full workflow
`survival_meta.md`	survmeta + KM pseudo-IPD reconstruction
`tsa_diagnostics.md`	TSA + Baujat + Drapery + Bootmeta
`diagnosis_meta.md`	mada::reitsma bivariate + SROC
`review_workflow.md`	metagear: PRISMA, screening, PDF batch, digitize
`data_templates.md`	Data input templates per type
`citations.md`	Full citation list
`r_packages.md`	Package details & installation
`advanced_api.md`	Reusable API reference
`svg_editing.md`	SVG editing tools & journal format conversion
`units.md`	Atomic task unit index (pipeline)
`language_policy.md`	Bilingual policy detailed companion
`report_template.md`	Report skeleton reference

Project Files

Changelog

Version / fix log → CHANGELOG.md.

meta-analysis-journal-selector

SCI journal selection assistant for meta-analysis and systematic review papers. Use this skill when the user has completed or is completing a meta-analysis a...

1 星标

Hcc Meta Ai

AI-assisted meta-analysis workflow for hepatocellular carcinoma comparative effectiveness research. Six-element prompt framework covering full pipeline from...

1 次安装

Clinical Trial Sample Size & Power / 临床试验样本量与检验效能专家

Sample size and power calculation tool for clinical trial practitioners. No local R install needed — a cloud R compute service covers all 49 test types and returns publication-grade SVG figures. Natural-language driven; full reproducible R code is returned by default; default output in Chinese or English per OS language setting (prompt can force-switch). / 为临床试验从业者提供的样本量与检验效能计算工具。本地无需安装 R，直接提供云端 R 计算服务（覆盖 49 种检验，并提供 SVG 出版级别图形）。自然语言驱动，默认回传完整 R 代码；默认按操作系统语言设定输出中文或英文（提示词可强制切换）。

unisound-clinical-trial-statistics

药企药物研发辅助临床试验数据统计。参考 Statistical Analysis skill 的 descriptive statistics 与 group comparison 部分，构建试验分析支持能力。

13 次安装

MSA分析技能

辅助完成测量系统分析(MSA)，支持Gage R&R、偏倚、线性、稳定性分析；当用户需要进行测量系统评估、了解测量系统准确性、生成MSA分析报告时使用