AGENT SKILL

github-actions-creator

Use when the user wants to create, generate, or set up a GitHub Actions workflow. Handles CI/CD pipelines, testing, deployment, linting, security scanning, release automation, Docker builds, scheduled tasks, and any custom workflow for any language or framework.

License 许可
MIT
最近更新
2026-11-11
Stars
★ 3.3w
Forks
3.9k

01 前言 Front Matter

name
github-actions-creator
description
Use when the user wants to create, generate, or set up a GitHub Actions workflow. Handles CI/CD pipelines, testing, deployment, linting, security scanning, release automation, Docker builds, scheduled tasks, and any custom workflow for any language or framework.

02 资源文件 5

03 SKILL.md 正文

GitHub Actions Creator

You are an expert at creating GitHub Actions workflows. When the user asks you to create a GitHub Action, follow this structured process to deliver a production-ready workflow file.

Workflow Creation Process

Step 1: Analyze the Project

Before writing any YAML, scan the project to understand the stack:

  1. Check for language/framework indicators:

    • package.json → Node.js (check for React, Next.js, Vue, Angular, Svelte, etc.)
    • requirements.txt / pyproject.toml / setup.py → Python
    • go.mod → Go
    • Cargo.toml → Rust
    • pom.xml / build.gradle → Java/Kotlin
    • Gemfile → Ruby
    • composer.json → PHP
    • pubspec.yaml → Dart/Flutter
    • Package.swift → Swift
    • *.csproj / *.sln → .NET
  2. Check for existing CI/CD:

    • .github/workflows/ → existing workflows (avoid conflicts)
    • Dockerfile → container builds available
    • docker-compose.yml → multi-service setup
    • vercel.json / netlify.toml → deployment targets
    • terraform/ / pulumi/ → infrastructure as code
  3. Check for tooling:

    • .eslintrc* / eslint.config.* → ESLint configured
    • prettier* → Prettier configured
    • jest.config* / vitest.config* / pytest.ini → test framework
    • .env.example → environment variables needed
    • Makefile → build commands available

Step 2: Ask Clarifying Questions (if needed)

If the user's request is ambiguous, ask ONE focused question. Common clarifications:

  • "Create a CI pipeline" → "Should it run tests only, or also lint and type-check?"
  • "Add deployment" → "Where does this deploy? (Vercel, AWS, GCP, Docker Hub, etc.)"
  • "Set up tests" → "Should tests run on PR only, or also on push to main?"

If the intent is clear, skip this step and proceed.

Step 3: Generate the Workflow

Create the .github/workflows/{name}.yml file following these rules:

File Naming

  • Use descriptive kebab-case names: ci.yml, deploy-production.yml, release.yml
  • For simple CI: ci.yml
  • For deployment: deploy.yml or deploy-{target}.yml
  • For scheduled tasks: scheduled-{task}.yml

YAML Structure Rules

name: Human-readable name        # Always include

on:                               # Use the most specific triggers
  push:
    branches: [main]              # Specify branches explicitly
    paths-ignore:                 # Skip docs-only changes when appropriate
      - '**.md'
      - 'docs/**'
  pull_request:
    branches: [main]

permissions:                      # Always set minimal permissions
  contents: read

concurrency:                      # Prevent duplicate runs on PRs
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: true

jobs:
  job-name:
    runs-on: ubuntu-latest        # Default to ubuntu-latest
    timeout-minutes: 15           # Always set a timeout
    steps:
      - uses: actions/checkout@v4 # Always pin to major version

Core Patterns by Use Case

CI (Test + Lint)

Trigger: pull_request + push to main Jobs: lint, test (parallel when possible) Key features: dependency caching, matrix testing for multiple versions

Deployment

Trigger: push to main (or release tags) Jobs: test → build → deploy (sequential with needs) Key features: environment protection, secrets for credentials, status checks

Release / Publish

Trigger: push tags matching v* or workflow_dispatch Jobs: test → build → publish → create GitHub Release Key features: changelog generation, artifact upload, npm/PyPI/Docker publish

Scheduled Tasks

Trigger: schedule with cron expression Jobs: single job with the task Key features: workflow_dispatch for manual trigger too, failure notifications

Security Scanning

Trigger: pull_request + schedule (weekly) Jobs: dependency audit, SAST, secret scanning Key features: SARIF upload to GitHub Security tab, fail on critical

Docker Build & Push

Trigger: push to main + tags Jobs: build → push to registry Key features: multi-platform builds, layer caching, image tagging strategy

Essential Actions Reference

Setup Actions (always pin to major version)

Action Purpose
actions/checkout@v4 Clone repository
actions/setup-node@v4 Node.js with caching
actions/setup-python@v5 Python with caching
actions/setup-go@v5 Go with caching
actions/setup-java@v4 Java/Kotlin
dtolnay/rust-toolchain@stable Rust toolchain
ruby/setup-ruby@v1 Ruby with bundler cache
actions/setup-dotnet@v4 .NET SDK

Build & Deploy Actions

Action Purpose
docker/build-push-action@v6 Docker multi-platform builds
docker/login-action@v3 Docker registry authentication
aws-actions/configure-aws-credentials@v4 AWS authentication
google-github-actions/auth@v2 GCP authentication
azure/login@v2 Azure authentication
cloudflare/wrangler-action@v3 Cloudflare Workers deploy
amondnet/vercel-action@v25 Vercel deployment

Quality & Security Actions

Action Purpose
github/codeql-action/analyze@v3 CodeQL SAST scanning
aquasecurity/trivy-action@master Container vulnerability scan
codecov/codecov-action@v4 Coverage upload
actions/dependency-review-action@v4 Dependency audit on PRs

Utility Actions

Action Purpose
actions/cache@v4 Generic caching
actions/upload-artifact@v4 Store build artifacts
actions/download-artifact@v4 Retrieve artifacts between jobs
softprops/action-gh-release@v2 Create GitHub Releases
slackapi/slack-github-action@v2 Slack notifications
peter-evans/create-pull-request@v7 Automated PR creation

Security Best Practices (ALWAYS follow)

  1. Minimal permissions: Always declare permissions at workflow or job level
  2. Pin actions to major version: Use @v4 not @main or full SHA for readability
  3. Never echo secrets: Secrets are masked but avoid echo ${{ secrets.X }}
  4. Use environments: For production deploys, use GitHub Environments with protection rules
  5. Validate inputs: For workflow_dispatch, validate input values
  6. Avoid script injection: Never use ${{ github.event.*.body }} directly in run: — pass via environment variables
  7. Use GITHUB_TOKEN: Prefer ${{ secrets.GITHUB_TOKEN }} over PATs when possible
  8. Concurrency controls: Use concurrency to prevent parallel deploys
# WRONG - script injection vulnerability
- run: echo "${{ github.event.issue.title }}"

# CORRECT - pass through environment variable
- run: echo "$ISSUE_TITLE"
  env:
    ISSUE_TITLE: ${{ github.event.issue.title }}

Caching Strategies

Node.js

- uses: actions/setup-node@v4
  with:
    node-version: 20
    cache: 'npm'  # or 'yarn' or 'pnpm'

Python

- uses: actions/setup-python@v5
  with:
    python-version: '3.12'
    cache: 'pip'  # or 'poetry' or 'pipenv'

Go

- uses: actions/setup-go@v5
  with:
    go-version: '1.22'
    cache: true

Rust

- uses: actions/cache@v4
  with:
    path: |
      ~/.cargo/bin/
      ~/.cargo/registry/index/
      ~/.cargo/registry/cache/
      target/
    key: ${{ runner.os }}-cargo-${{ hashFiles('**/Cargo.lock') }}

Docker

- uses: docker/build-push-action@v6
  with:
    cache-from: type=gha
    cache-to: type=gha,mode=max

Matrix Testing Patterns

Multiple Node.js versions

strategy:
  matrix:
    node-version: [18, 20, 22]
  fail-fast: false

Multiple OS

strategy:
  matrix:
    os: [ubuntu-latest, macos-latest, windows-latest]
runs-on: ${{ matrix.os }}

Complex matrix with exclusions

strategy:
  matrix:
    os: [ubuntu-latest, windows-latest]
    node-version: [18, 20]
    exclude:
      - os: windows-latest
        node-version: 18

Cron Syntax Quick Reference

Schedule Cron
Every hour 0 * * * *
Daily at midnight UTC 0 0 * * *
Weekdays at 9am UTC 0 9 * * 1-5
Weekly on Sunday 0 0 * * 0
Monthly 1st 0 0 1 * *

Output Format

After creating the workflow file, provide:

  1. What the workflow does — one-paragraph summary
  2. Required secrets — list any secrets the user needs to configure in Settings > Secrets
  3. Required permissions — if the workflow needs non-default repository permissions
  4. How to test — how to trigger the workflow (push, create PR, manual dispatch)

Common Patterns to Combine

When the user asks for something generic like "set up CI/CD", create a single workflow with multiple jobs:

jobs:
  lint:        # Fast feedback
  test:        # Core validation
  build:       # Ensure it compiles/bundles
    needs: [lint, test]
  deploy:      # Only after everything passes
    needs: build
    if: github.ref == 'refs/heads/main'

Keep workflows focused. Prefer one workflow per concern over one massive workflow, unless the jobs are tightly coupled.

04 同仓库更多 Skills

davila7
davila7/claude-code-templates

wealthville-pool-scoring

Score DeFi liquidity pools before providing liquidity. Returns a 0-100 pool score and an Enter/Hold/Exit/Reduce/Avoid verdict with calibrated confidence for ~68,800 Solana pools and 575 EVM pools, plus a public miss-inclusive track record. Use when asked whether a pool is worth LPing into, which pools look best right now, or how accurate the scores have actually been.

Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

technical-debt-auditor

Scans a codebase and produces a quantified technical debt report with a prioritized remediation plan. Analyzes dead code, TODO/FIXME/HACK markers, dependency staleness, complexity hotspots, code duplication signals, test coverage gaps, unused exports, and outdated patterns. Use this skill when the user asks to audit their codebase for tech debt, wants a health check of their code quality, says things like "where should I focus refactoring", "find code smells", "audit my codebase", "what technical debt do we have", "prioritize cleanup work", or asks about code quality metrics. Also trigger when the user mentions "code rot", "legacy cleanup", "refactoring priorities", or wants to understand the overall health of a project before a major effort. Trigger even if the user doesn't say "technical debt" explicitly — any request to systematically assess code quality and identify improvement priorities belongs here.

Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

readonly-sql-guard

Make an agent's database access provably read-only. Use when an agent runs caller-supplied SQL against a production/ERP/analytics database (a "query" MCP tool, text-to-SQL, an admin console), or when reviewing DB-access code for write/DDL/side-effect/transaction-escape risks. Checks what a statement CALLS, not just its shape.

数据分析师后端与接口Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

web-artifacts-builder

Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.

前端开发
★ 3.3w
davila7
davila7/claude-code-templates

webapp-testing

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

测试与质量前端开发
★ 3.3w
davila7
davila7/claude-code-templates

discernment-nudge

After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.

内容创作者写作与文案Skill与智能体开发
★ 3.3w