关于
This skill installs and configures the `putior` R package for creating workflow visualizations from code annotations. It handles installation from CRAN or GitHub, sets up optional dependencies, and verifies the complete annotation-to-diagram pipeline. Use it for initial setup, environment preparation, or when a downstream skill requires `putior`.
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/install-putior在 Claude Code 中复制并粘贴此命令以安装该技能
技能文档
Install putior
Install putior R package + optional dependencies so annotation-to-diagram pipeline ready to use.
When Use
- Setting up putior for first time in project or environment
- Preparing machine for workflow visualization tasks
- Downstream skill (analyze-codebase-workflow, generate-workflow-diagram) needs putior installed
- Restoring environment after R version upgrade or renv wipe
Inputs
- Required: Access to R installation (>= 4.1.0)
- Optional: Install from CRAN (default) or GitHub dev version
- Optional: Which optional dependency groups to install: MCP (
mcptools,ellmer), interactive (shiny,shinyAce), logging (logger), ACP (plumber2)
Steps
Step 1: Verify R Installation
Confirm R available + meets minimum version requirement.
R.Version()$version.string
# Must be >= 4.1.0
# From WSL with Windows R
"/mnt/c/Program Files/R/R-4.5.2/bin/Rscript.exe" -e "cat(R.version.string)"
Got: R version string printed, >= 4.1.0.
If fail: Install or upgrade R. On Windows, download from https://cran.r-project.org/bin/windows/base/. On Linux, use sudo apt install r-base.
Step 2: Install putior
Install from CRAN (stable) or GitHub (dev).
# CRAN (recommended)
install.packages("putior")
# GitHub dev version (if latest features needed)
remotes::install_github("pjt222/putior")
Got: Package installs no errors. library(putior) loads silently.
If fail: CRAN installation fails with "not available for this version of R"? Use GitHub version. GitHub fails? Check remotes installed: install.packages("remotes").
Step 3: Install Optional Dependencies
Install optional packages based on required functionality.
# MCP server integration (for AI assistant access)
remotes::install_github("posit-dev/mcptools")
install.packages("ellmer")
# Interactive sandbox
install.packages("shiny")
install.packages("shinyAce")
# Structured logging
install.packages("logger")
# ACP server (agent-to-agent communication)
install.packages("plumber2")
Got: Each package installs no errors.
If fail: mcptools → ensure remotes installed first. System dependency errors on Linux? Install required libraries (sudo apt install libcurl4-openssl-dev for httr2 dependency).
Step 4: Verify Installation
Run basic pipeline to confirm everything works.
library(putior)
# Check package version
packageVersion("putior")
# Verify core functions are available
stopifnot(
is.function(put),
is.function(put_auto),
is.function(put_diagram),
is.function(put_generate),
is.function(put_merge),
is.function(put_theme)
)
# Test basic pipeline with a temp file
tmp <- tempfile(fileext = ".R")
writeLines("# put id:'test', label:'Hello putior'", tmp)
cat(put_diagram(put(tmp)))
Got: Mermaid flowchart code printed to console containing test + Hello putior.
Key defaults: All scan functions (
put(),put_auto(),put_generate(),put_merge()) default torecursive = TRUE, scanning subdirectories automatic. Breaking change from pre-0.2.0 versions where default wasFALSE. All scan functions also acceptexcludeparameter for regex-based file filtering (put("./src/", exclude = "test_")).
Optional shiny package installed? Try interactive sandbox:
putior::run_sandbox()
Launches browser-based editor where you experiment with PUT annotation syntax + see diagrams rendered real time.
If fail: put not found? Package didn't install correctly. Reinstall with install.packages("putior", dependencies = TRUE). Diagram empty? Verify temp file created + annotation syntax uses single quotes inside double quotes.
Checks
-
library(putior)loads no errors -
packageVersion("putior")returns valid version -
put()with file containing valid PUT annotation returns data frame with one row -
put_diagram()produces Mermaid code starting withflowchart - All requested optional dependencies load no errors
Pitfalls
- Wrong quote nesting: PUT annotations use single quotes inside annotation:
id:'name', notid:"name"(conflicts with comment string delimiter in some contexts). - Missing Pandoc for vignettes: Plan to build putior vignettes locally? Ensure
RSTUDIO_PANDOCset in.Renviron. - renv isolation: Project uses renv? Must install putior inside renv library. Run
renv::install("putior")notinstall.packages("putior"). - GitHub rate limits: Installing
mcptoolsfrom GitHub may fail withoutGITHUB_PAT. Set one viausethis::create_github_token().
See Also
analyze-codebase-workflow— next step after installation to survey codebaseconfigure-putior-mcp— set up MCP server after installing optional depsmanage-renv-dependencies— manage putior within renv environmentconfigure-mcp-server— general MCP server configuration
GitHub 仓库
Frequently asked questions
What is the install-putior skill?
install-putior is a Claude Skill by pjt222. Skills package instructions and resources that Claude loads on demand, so Claude can perform install-putior-related tasks without extra prompting.
How do I install install-putior?
Use the install commands on this page: add install-putior 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 install-putior belong to?
install-putior is in the Meta category, tagged ai, api, mcp, automation and design.
Is install-putior free to use?
Yes. install-putior 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。
