关于
This skill sets up an R project structure compliant with GxP regulations like 21 CFR Part 11. It establishes a validated environment with documentation, change control, and electronic records handling for regulatory submissions. Use it when starting R analysis projects in regulated pharma, biotech, or clinical trial environments.
快速安装
Claude Code
推荐npx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/setup-gxp-r-project在 Claude Code 中复制并粘贴此命令以安装该技能
技能文档
設 GxP R 項
建合 GxP 規之 R 項結構為驗算。
用
- 規境(藥、生技、醫器)始 R 析項→用
- 為臨試析設 R→用
- 為規呈建驗算境→用
- 行 21 CFR Part 11 或 EU Annex 11 需→用
入
- 必:項範與規框(FDA、EMA、二)
- 必:所驗 R 本與包本
- 必:驗策(險導法)
- 可:既電腦系 SOP
- 可:質管系接需
行
一:建驗項結構
gxp-project/
├── R/
│ ├── 01_data_import.R
│ ├── 02_data_processing.R
│ └── 03_analysis.R
├── validation/
│ ├── validation_plan.md
│ ├── risk_assessment.md
│ ├── iq/
│ │ ├── iq_protocol.md
│ │ └── iq_report.md
│ ├── oq/
│ │ ├── oq_protocol.md
│ │ └── oq_report.md
│ ├── pq/
│ │ ├── pq_protocol.md
│ │ └── pq_report.md
│ └── traceability_matrix.md
├── tests/
│ ├── testthat.R
│ └── testthat/
│ ├── test-data_import.R
│ └── test-analysis.R
├── data/
│ ├── raw/
│ └── derived/
├── output/
├── docs/
│ ├── sop_references.md
│ └── change_log.md
├── renv.lock
├── DESCRIPTION
├── .Rprofile
└── CLAUDE.md
得:完目結構含 R/、validation/(含 iq/、oq/、pq/)、tests/testthat/、data/raw/、data/derived/、output/、docs/。
敗:目缺→mkdir -p 建。驗於正項根。既項僅建缺目、勿覆既結構。
二:建驗計
建 validation/validation_plan.md:
# Validation Plan
## 1. Purpose
This plan defines the validation strategy for [Project Name] using R [version].
## 2. Scope
- R version: 4.5.0
- Packages: [list with versions]
- Analysis: [description]
- Regulatory framework: 21 CFR Part 11 / EU Annex 11
## 3. Risk Assessment Approach
Using GAMP 5 risk-based categories:
- Category 3: Non-configured products (R base)
- Category 4: Configured products (R packages with default settings)
- Category 5: Custom applications (custom R scripts)
## 4. Validation Activities
| Activity | Category 3 | Category 4 | Category 5 |
|----------|-----------|-----------|-----------|
| IQ | Required | Required | Required |
| OQ | Reduced | Standard | Enhanced |
| PQ | N/A | Standard | Enhanced |
## 5. Roles and Responsibilities
- Validation Lead: [Name]
- Developer: [Name]
- QA Reviewer: [Name]
- Approver: [Name]
## 6. Acceptance Criteria
All tests must pass with documented evidence.
得:validation/validation_plan.md 完含範、GAMP 5 險類、驗動陣、職責、接準。引特 R 本與規框。
敗:規框不明→諮組 QA 部為適 SOP。計未審准前勿進驗動。
三:用 renv 鎖依
renv::init()
renv::install("[email protected]")
renv::install("[email protected]")
renv::snapshot()
renv.lock 為控包冊。
得:renv.lock 存含諸需包準本。renv::status() 報無問題。每包本釘(如 [email protected])非浮。
敗:renv::install() 特本敗→察本存於 CRAN 檔。用 renv::install("package@version", repos = "https://packagemanager.posit.co/cran/latest") 為檔本。
四:行版控
git init
git add .
git commit -m "Initial validated project structure"
git config user.signingkey YOUR_GPG_KEY
git config commit.gpgsign true
得:項於 git 控含署名提啟。首提含驗結構與 renv.lock。
敗:GPG 署敗→gpg --list-secret-keys 驗 GPG 鑰配。無 GPG 境→文錄偏、用無署提含 docs/change_log.md 之手審跡。
五:建 IQ 議
validation/iq/iq_protocol.md:
# Installation Qualification Protocol
## Objective
Verify that R and required packages are correctly installed.
## Test Cases
### IQ-001: R Version Verification
- **Requirement**: R 4.5.0 installed
- **Procedure**: Execute `R.version.string`
- **Expected:** "R version 4.5.0 (date)"
- **Result**: [ PASS / FAIL ]
### IQ-002: Package Installation Verification
- **Requirement**: All packages in renv.lock installed
- **Procedure**: Execute `renv::status()`
- **Expected:** "No issues found"
- **Result**: [ PASS / FAIL ]
### IQ-003: Package Version Verification
- **Procedure**: Execute `installed.packages()[, c("Package", "Version")]`
- **Expected:** Versions match renv.lock exactly
- **Result**: [ PASS / FAIL ]
得:validation/iq/iq_protocol.md 含 R 本驗、包裝驗、包本驗之測例、各含明期果與過/敗欄。
敗:IQ 議板不合組 SOP→保需欄(需、程、期果、實果、過/敗)改式。諮 QA 為准板。
六:書自動 OQ/PQ 測
test_that("primary analysis produces validated results", {
test_data <- read.csv(test_path("fixtures", "validation_dataset.csv"))
result <- primary_analysis(test_data)
expect_equal(result$estimate, 2.345, tolerance = 1e-3)
expect_equal(result$p_value, 0.012, tolerance = 1e-3)
expect_equal(result$ci_lower, 1.234, tolerance = 1e-3)
})
得:自測檔於 tests/testthat/ 覆 OQ(各函操驗)與 PQ(端對端對獨算參值驗)。測用顯數忍。
敗:獨算之參值未備(如 SAS)→建占測含 skip("Awaiting independent reference values") 並文錄於追陣。
七:建追陣
# Traceability Matrix
| Req ID | Requirement | Test ID | Test Description | Status |
|--------|-------------|---------|------------------|--------|
| REQ-001 | Import CSV data correctly | OQ-001 | Verify data dimensions and types | PASS |
| REQ-002 | Calculate primary endpoint | PQ-001 | Compare against reference results | PASS |
| REQ-003 | Generate report output | PQ-002 | Verify report contains all sections | PASS |
得:validation/traceability_matrix.md 連各需於 ≥ 1 測例、各測例連於需。無孤需或測。
敗:需未測→建測例或文錄險導排除。測無連需→連既需或除為範外。
驗
- 項結構循文錄板
- renv.lock 含諸依準本
- 驗計完准
- IQ 議成行
- OQ 測例覆諸配功
- PQ 測對獨算果驗
- 追陣連需於測
- 變控程文錄
忌
install.packages()無本釘:恆用 renv 含鎖本- 缺審跡:諸變必文錄。用 git 署提
- 過驗:施險導法。非每 CRAN 包需類 5 驗
- 忘系級驗:OS 與 R 裝亦需 IQ
- 無獨驗:PQ 宜對獨算果(SAS、手算)較
參
write-validation-documentationimplement-audit-trailvalidate-statistical-outputmanage-renv-dependencies
GitHub 仓库
Frequently asked questions
What is the setup-gxp-r-project skill?
setup-gxp-r-project is a Claude Skill by pjt222. Skills package instructions and resources that Claude loads on demand, so Claude can perform setup-gxp-r-project-related tasks without extra prompting.
How do I install setup-gxp-r-project?
Use the install commands on this page: add setup-gxp-r-project to Claude Code as a plugin, or clone its repository into your skills directory, then restart Claude so it picks up the skill.
What category does setup-gxp-r-project belong to?
setup-gxp-r-project is in the Meta category, tagged word and design.
Is setup-gxp-r-project free to use?
Yes. setup-gxp-r-project is listed on AIMCP and free to install. It runs inside Claude, so no separate service account is required to use the skill itself.
相关推荐技能
Content Collections 是一个 TypeScript 优先的构建工具,可将本地 Markdown/MDX 文件转换为类型安全的数据集合。它专为构建博客、文档站和内容密集型 Vite+React 应用而设计,提供基于 Zod 的自动模式验证。该工具涵盖从 Vite 插件配置、MDX 编译到生产环境部署的完整工作流。
这个Claude Skill为开发者提供完整的Polymarket预测市场开发支持,涵盖API调用、交易执行和市场数据分析。关键特性包括实时WebSocket数据流,可监控实时交易、订单和市场动态。开发者可用它构建预测市场应用、实施交易策略并集成实时市场预测功能。
该Skill帮助开发者创建OpenCode插件,用于接入命令、文件、LSP等25+种事件。它提供了插件结构、事件API规范和JavaScript/TypeScript实现模式,适合需要拦截操作、扩展功能或自定义事件处理的场景。开发者可通过它快速构建响应式模块来增强OpenCode AI助手的能力。
SGLang是一个专为LLM设计的高性能推理框架,特别适用于需要结构化输出的场景。它通过RadixAttention前缀缓存技术,在处理JSON、正则表达式、工具调用等具有重复前缀的复杂工作流时,能实现极速生成。如果你正在构建智能体或多轮对话系统,并追求远超vLLM的推理性能,SGLang是理想选择。
