By walmanarias
Spec-Driven Development lifecycle with TDD and E2E testing. Ships lifecycle agents (spec, tdd, implement, e2e, review, refactor, architect) and stack experts (React, React Native, Node/Express/Fastify, NestJS, Next.js, C#/ASP.NET Core, Avalonia/XAML desktop, PostgreSQL, MongoDB), reusable skills, slash commands, and enforcing quality-gate hooks for clean, maintainable, scalable code. Cross-platform incl. macOS Apple Silicon.
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Generate or update a full GitHub Actions CI/CD pipeline (build → test → deploy) for this project.
Add reliable end-to-end / integration tests for a flow.
Run the full Spec-Driven Development + TDD loop for a feature, end to end.
Make failing tests pass with minimal clean code (GREEN -> REFACTOR).
Produce an architecture/design brief and ADRs before specifying or coding.
Use BEFORE writing a spec or any code for a non-trivial feature or a new service/module. Produces a system-design brief — boundaries, data model, API contracts, key trade-offs, and chosen patterns — that the spec-writer and implementers build on. Trigger on "design", "architecture", "how should we structure", new service, or cross-cutting change.
Use for cross-platform desktop UI with Avalonia and XAML (AXAML) in C# — views, MVVM, data binding, styles/themes, custom controls, and Avalonia testing (headless xUnit + Appium). Triggers on "Avalonia", "XAML", "AXAML", "desktop app", "MVVM", "view model". Targets modern .NET on macOS Apple Silicon. For ASP.NET Core APIs use dotnet-expert.
Use to set up or change a project's CI/CD — generate and maintain GitHub Actions pipelines that build, test, and deploy. Triggers on "ci/cd", "pipeline", "deploy", "github actions", "release", or a deploy target (Cloudflare, AWS, Azure, GCP, Railway, Fly.io, Vercel, Netlify, EAS, npm, NuGet). Defaults to staging → manual-approval → production.
Use after implementation or before opening/merging a PR to review a diff for clean code, correctness, security, performance, and test quality. Proactively invoke after any non-trivial change. Trigger on "review", "code review", "PR", or when work is ready to ship. Reviews; does not rewrite unless asked.
Use for data-layer work on PostgreSQL or MongoDB — schema/collection design, indexes, queries, migrations, transactions, and ORM/ODM usage (Prisma/TypeORM/Drizzle, Mongoose). Triggers on "schema", "migration", "index", "query performance", "N+1", "Postgres", "Mongo", "transaction". Designs and reviews the persistence layer.
Use when setting up or changing a project's CI/CD — generating GitHub Actions pipelines that build, test, and deploy. Trigger on "ci/cd", "pipeline", "deploy", "github actions", "release", "staging", "production", or a deploy target (Cloudflare, AWS, Azure, GCP, Railway, Fly.io, Vercel, Netlify, EAS, npm, NuGet).
Use when writing, reviewing, or refactoring code to apply clean-code, SOLID, and maintainability standards for TypeScript/JavaScript. Trigger on "clean code", "refactor", "code smell", "maintainable", "SOLID", "review", or whenever assessing whether code is well-written.
Use when adding or fixing end-to-end / integration tests, to select the right tool for the stack and write reliable, hermetic flows. Trigger on "e2e", "end-to-end", "integration test", "Playwright", "Detox", "Supertest", "flaky test", or before shipping a user-facing flow.
Use when writing or updating a pull request title and description from a branch's commits. Produces a scannable, stack-aware PR body — emoji-tagged sections added only when the commits warrant them — while preserving existing screenshots, videos, and designer feedback. Trigger on "PR description", "update PR", "write the PR", or "/update-pr".
Use when starting any non-trivial feature or change, to run the Spec-Driven Development loop — write a testable spec, derive tests, implement, and verify against the spec. Trigger on "spec", "SDD", "new feature", "requirements", or when work should start from a written contract rather than ad-hoc coding.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
A reusable Claude Code workflow that enforces Spec-Driven Development (SDD) and Test-Driven Development (TDD) across a full-stack TypeScript/JavaScript, C#/.NET, and Python stack: React, React Native, Node.js (Express/Fastify), NestJS, Next.js, C# / ASP.NET Core Web APIs, Avalonia (XAML) desktop, Python (Django/DRF, FastAPI, Flask), PostgreSQL, MongoDB.
It ships as a Claude Code plugin (agentic-sdd) and as a copyable .claude/ folder, so you can install it in any repo and start shipping clean, tested, maintainable code. Cross-platform, including macOS (Tahoe) on Apple Silicon — .NET work targets modern cross-platform .NET 8/9 (arm64-native), not the Windows-only .NET Framework.
Read
CLAUDE.mdfor the full source of truth on how the agents, skills, commands, hooks, and rules work together.
architect, spec-writer, tdd-test-writer, implementer, e2e-tester, code-reviewer, refactorer, cicd-engineer) + 11 stack experts (React, React Native, Node/Express/Fastify, NestJS, Next.js, C#/ASP.NET Core, Avalonia/XAML, Django/DRF, FastAPI, Flask, PostgreSQL/MongoDB)./feature, /plan, /spec, /tdd, /implement, /e2e, /review, /refactor, /cicd, /update-pr, /triage-copilot, /triage-reviews, /ship.spec-driven-development, tdd-workflow, e2e-testing, clean-code, stack-testing-recipes, cicd-pipelines, pr-description.debugger/Debugger.Break()/breakpoint(), format/lint changed files (ESLint, dotnet format, or ruff), and gate git commit on the toolchains the repo has (JS: lint + types + tests; .NET: format + build -warnaserror + test; Python: ruff + mypy + pytest).--with-ci. This repo's own CI just validates the plugin.Open this folder with Claude and run:
/feature add a health-check endpoint that returns build version and DB status
It runs the whole loop — spec → failing tests → implementation → E2E → review → ship — stopping at the approval gates.
Two ways to install. Option A (plugin) is recommended — it updates cleanly from this repo and copies nothing into your target's git history. Option B (copy) is fully self-contained — use it when you can't depend on a marketplace or want the workflow checked into the repo. After either, open the target repo with Claude and run /feature … to start.
One-time setup: push this template to a GitHub repo you control. It already has .claude-plugin/marketplace.json at the root, which is what makes the repo a plugin marketplace.
Then, in the target project:
/plugin marketplace add walmanarias/agentic-workflow
/plugin install agentic-sdd@agentic-workflow
walmanarias/agentic-workflow with your repo path. A full Git URL or a local path also work — handy during local development before you push: /plugin marketplace add ./path/to/agentic-workflow.--scope project to scope it to this repo only, or --scope local for a personal, uncommitted install: /plugin install agentic-sdd@agentic-workflow --scope project./plugin anytime for the interactive menu (Discover / Installed / Marketplaces / Errors).Make it automatic for a team. Commit these keys to the target repo's .claude/settings.json, and anyone who opens the repo with Claude gets the marketplace and plugin without manual steps:
{
"extraKnownMarketplaces": {
"agentic-workflow": {
"source": { "source": "github", "repo": "walmanarias/agentic-workflow" }
}
},
"enabledPlugins": {
"agentic-sdd@agentic-workflow": true
}
}
Updating an existing install — pull the latest after this template changes:
/plugin marketplace update agentic-workflow # refresh the catalog from your repo
/plugin install agentic-sdd@agentic-workflow # reinstall to pull the newest version
Or enable auto-update from /plugin → Marketplaces. To remove: /plugin uninstall agentic-sdd@agentic-workflow (and /plugin marketplace remove agentic-workflow to drop the source — note that also uninstalls its plugins).
.claude/ folder (self-contained)From your local checkout of this template:
bash scripts/install.sh /path/to/your/repo
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimnpx claudepluginhub walmanarias/agentic-workflow --plugin agentic-sddComprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Lazy senior dev mode. Forces the simplest, shortest solution that actually works: YAGNI, stdlib first, no unrequested abstractions.