MCP HubMCP Hub
스킬 목록으로 돌아가기

setup-putior-ci

pjt222
업데이트됨 6 days ago
14 조회
17
2
17
GitHub에서 보기
메타automation

정보

이 Claude Skill은 코드가 푸시될 때마다 자동으로 putior 워크플로 다이어그램을 생성하고 업데이트하는 GitHub Actions CI/CD 파이프라인을 설정합니다. 워크플로 YAML, 센티넬 마커가 포함된 R 스크립트 생성, 업데이트된 다이어그램 자동 커밋(README 통합을 통한 제자리 업데이트 포함)을 처리합니다. 다이어그램이 항상 최신 코드를 반영하도록 보장하기 위해 사용하며, 특히 다중 기여자 환경이나 수동 재생성을 자동화할 때 유용합니다.

빠른 설치

Claude Code

추천
기본
npx skills add pjt222/agent-almanac -a claude-code
플러그인 명령대체
/plugin add https://github.com/pjt222/agent-almanac
Git 클론대체
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/setup-putior-ci

Claude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요

문서

Set Up putior CI/CD

Configure GitHub Actions to automatically regenerate workflow diagrams when source code changes, keeping documentation in sync with code.

When to Use

  • Workflow diagrams should always reflect the current state of the code
  • The project has CI/CD and wants automated documentation updates
  • Multiple contributors may change workflow-affecting code
  • Replacing manual diagram regeneration with automated pipeline

Inputs

  • Required: GitHub repository with putior annotations in source files
  • Required: Target file for diagram output (e.g., README.md, docs/workflow.md)
  • Optional: putior theme (default: "github")
  • Optional: Source directories to scan (default: "./R/" or "./src/")
  • Optional: Branch to trigger on (default: main)

Procedure

Step 1: Create GitHub Actions Workflow

Create the workflow YAML file for automated diagram generation.

# .github/workflows/update-workflow-diagram.yml
name: Update Workflow Diagram
locale: wenyan-ultra
source_locale: en
source_commit: 82c77053
translator: "Julius Brussee homage — caveman"
translation_date: "2026-04-19"

on:
  push:
    branches: [main]
    paths:
      - 'R/**'
      - 'src/**'
      - 'scripts/**'

permissions:
  contents: write

jobs:
  update-diagram:
    if: github.actor != 'github-actions[bot]'
    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v4

      - uses: r-lib/actions/setup-r@v2
        with:
          use-public-rspm: true

      - name: Install putior
        run: |
          install.packages("putior")
        shell: Rscript {0}

      - name: Generate workflow diagram
        run: |
          Rscript scripts/generate-workflow-diagram.R

      - name: Commit updated diagram
        run: |
          git config --local user.name "github-actions[bot]"
          git config --local user.email "github-actions[bot]@users.noreply.github.com"
          git add README.md docs/workflow.md  # Adjust to match your target files
          git diff --staged --quiet || git commit -m "docs: update workflow diagram [skip ci]"
          git push

Expected: File created at .github/workflows/update-workflow-diagram.yml.

On failure: Ensure the .github/workflows/ directory exists. Adjust the paths filter to match where annotated source files live in the repository.

Step 2: Write Generation Script

Create the R script that generates the diagram and updates target files using sentinel markers.

# scripts/generate-workflow-diagram.R
library(putior)

# Scan source files for annotations (exclude build scripts to avoid circular refs)
workflow <- put_merge("./R/", merge_strategy = "supplement",
  exclude = c("generate-workflow-diagram\\.R$"),
  log_level = NULL)  # Set to "DEBUG" to troubleshoot CI diagram generation

# Generate Mermaid code
mermaid_code <- put_diagram(workflow, output = "raw", theme = "github")

# Read target file (e.g., README.md)
readme <- readLines("README.md")

# Find sentinel markers
start_marker <- "<!-- PUTIOR-WORKFLOW-START -->"
end_marker <- "<!-- PUTIOR-WORKFLOW-END -->"

start_idx <- which(readme == start_marker)
end_idx <- which(readme == end_marker)

if (length(start_idx) == 1 && length(end_idx) == 1 && end_idx > start_idx) {
  # Replace content between sentinels
  new_content <- c(
    readme[1:start_idx],
    "",
    "```mermaid",
    mermaid_code,
    "```",
    "",
    readme[end_idx:length(readme)]
  )
  writeLines(new_content, "README.md")
  cat("Updated README.md workflow diagram\n")
} else {
  warning("Sentinel markers not found in README.md. Add them manually:\n",
          start_marker, "\n", end_marker)
}

# Also write standalone diagram file
writeLines(
  c("# Workflow Diagram", "",
    "```mermaid", mermaid_code, "```"),
  "docs/workflow.md"
)
cat("Updated docs/workflow.md\n")

Expected: Script at scripts/generate-workflow-diagram.R that reads annotations, generates Mermaid code, and replaces content between sentinel markers.

On failure: If put_merge() returns empty, check that source paths match the repository layout. Adjust "./R/" to the actual source directory.

Step 3: Configure Auto-Commit

The workflow must avoid infinite loops where an auto-commit re-triggers the same workflow. Pushes made with the default GITHUB_TOKEN typically do not trigger new workflow runs, but the workflow also includes an explicit if: guard on the job as a safety net.

Key configuration points:

  • permissions: contents: write grants push access
  • if: github.actor != 'github-actions[bot]' skips the job when the push came from the bot itself
  • git diff --staged --quiet || git commit only commits if there are changes
  • [skip ci] in the commit message is a convention some CI systems honor (not built into GitHub Actions, but useful as a signal)
  • Bot identity used for commits: github-actions[bot]

Expected: The workflow only commits when diagrams actually change. No empty commits, no infinite loops.

On failure: If push fails with permission denied, check repository settings: Settings > Actions > General > Workflow permissions must be set to "Read and write permissions".

Step 4: Add Sentinel Markers to README

Insert sentinel markers in the target file where the diagram should appear.

## Workflow

<!-- PUTIOR-WORKFLOW-START -->
<!-- This section is auto-generated by putior CI. Do not edit manually. -->

```mermaid
flowchart TD
  A["Placeholder — will be replaced on next CI run"]
<!-- PUTIOR-WORKFLOW-END -->

**Expected:** Sentinel markers in README.md (or other target file). The content between them will be replaced on each CI run.

**On failure:** Ensure markers are on their own lines with no leading/trailing whitespace. The script matches exact line content.

### Step 5: Test the Pipeline

Trigger the workflow and verify the diagram updates.

```bash
# Make a small change to trigger the workflow
echo "# test" >> R/some-file.R
git add R/some-file.R
git commit -m "test: trigger workflow diagram update"
git push

# Monitor the GitHub Actions run
gh run watch

# Verify the diagram was updated
git pull
cat README.md | grep -A 5 "PUTIOR-WORKFLOW-START"

Expected: GitHub Actions run completes successfully. The diagram between sentinel markers in README.md is updated with current workflow data.

On failure: Check the Actions log for errors. Common issues:

  • putior package not available: add to DESCRIPTION Suggests or install explicitly in the workflow
  • Source path wrong: the R script's put_merge() path must be relative to the repo root
  • No sentinel markers: the script warns but doesn't crash; add markers to README.md

Validation

  • .github/workflows/update-workflow-diagram.yml exists and is valid YAML
  • scripts/generate-workflow-diagram.R runs without errors locally
  • README.md contains <!-- PUTIOR-WORKFLOW-START --> and <!-- PUTIOR-WORKFLOW-END --> sentinels
  • GitHub Actions workflow triggers on push to the correct branch and paths
  • Diagram content between sentinels is updated after a workflow run
  • Job-level if: guard prevents infinite commit loops from bot pushes
  • No changes = no commit (idempotent)

Common Pitfalls

  • Infinite CI loops: Pushes with the default GITHUB_TOKEN typically don't trigger new runs, but always add an explicit if: github.actor != 'github-actions[bot]' guard on the job. The [skip ci] tag in the commit message is a useful convention but is not a built-in GitHub Actions mechanism.
  • Permission denied on push: GitHub Actions needs write permission. Set permissions: contents: write in the workflow file, or configure it in repository settings.
  • Sentinel marker mismatch: If markers have trailing spaces, leading tabs, or are on the same line as other content, the script won't find them. Keep markers on their own clean lines.
  • Source path mismatch: The R script runs from the repo root. Paths like "./R/" or "./src/" must match the actual directory structure.
  • Package installation in CI: If the project uses renv, the CI workflow needs renv::restore() before putior is available. Alternatively, install putior explicitly in the workflow.
  • Large repos slowing CI: For repos with many source files, limit the paths trigger filter to directories that contain PUT annotations, not the entire repo.

Related Skills

  • generate-workflow-diagram — the manual version of what this CI automates
  • setup-github-actions-ci — general GitHub Actions CI/CD setup for R packages
  • build-ci-cd-pipeline — broader CI/CD pipeline design
  • annotate-source-files — annotations must exist before CI can generate diagrams
  • commit-changes — understanding auto-commit patterns

GitHub 저장소

pjt222/agent-almanac
경로: i18n/wenyan-ultra/skills/setup-putior-ci
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

연관 스킬

content-collections

메타

이 스킬은 콘텐츠 콜렉션(Content Collections)을 위한 프로덕션 검증된 설정을 제공합니다. 콘텐츠 콜렉션은 Markdown/MDX 파일을 Zod 검증이 포함된 타입 안전한 데이터 콜렉션으로 변환해주는 TypeScript 최우선 도구입니다. 블로그, 문서 사이트 또는 콘텐츠 중심의 Vite + React 애플리케이션을 구축할 때 타입 안전성과 자동 콘텐츠 검증을 보장하기 위해 사용하세요. Vite 플러그인 구성과 MDX 컴파일부터 배포 최적화 및 스키마 검증에 이르기까지 모든 것을 다룹니다.

스킬 보기

polymarket

메타

이 스킬은 개발자들이 Polymarket 예측 시장 플랫폼을 활용한 애플리케이션을 구축할 수 있도록 지원하며, 거래 및 시장 데이터를 위한 API 통합 기능을 포함합니다. 또한 WebSocket을 통한 실시간 데이터 스트리밍을 제공하여 실시간 거래와 시장 활동을 모니터링할 수 있습니다. 이를 통해 거래 전략을 구현하거나 실시간 시장 업데이트를 처리하는 도구를 생성하는 데 활용할 수 있습니다.

스킬 보기

creating-opencode-plugins

메타

이 스킬은 개발자들이 명령어, 파일, LSP 작업 등 25개 이상의 이벤트 유형에 연결되는 OpenCode 플러그인을 만들 수 있도록 돕습니다. JavaScript/TypeScript 모듈을 위한 플러그인 구조, 이벤트 API 명세, 구현 패턴을 제공합니다. OpenCode AI 어시스턴트의 라이프사이클을 사용자 정의 이벤트 기반 로직으로 가로채거나, 모니터링하거나, 확장해야 할 때 사용하세요.

스킬 보기

sglang

메타

SGLang은 RadixAttention 프리픽스 캐싱을 활용하여 JSON, 정규식, 에이전트 워크플로우를 위한 고속 구조화 생성에 특화된 고성능 LLM 서빙 프레임워크입니다. 특히 반복되는 프리픽스가 있는 작업에서 상당히 빠른 추론 속도를 제공하여 복잡한 구조화 출력 및 다중 턴 대화에 이상적입니다. 제약 디코딩이 필요하거나 광범위한 프리픽스 공유가 있는 애플리케이션을 구축할 때는 vLLM과 같은 대안보다 SGLang을 선택하십시오.

스킬 보기