@codewithmukesh/dotnet-init

@codewithmukesh/dotnet-init — AI coding skill

View in AI SkillSafe app
0 downloads
0 stars
0 demos
SKILL.md
namedotnet-init
description>

/dotnet-init

What

Interactively initializes a .NET project for use with dotnet-claude-kit. Detects the project type, asks targeted questions about architecture and tech stack, then generates a fully customized CLAUDE.md in the project root.

No manual template copying required.

When

  • Starting a new .NET project with Claude Code
  • Adding dotnet-claude-kit to an existing project
  • "init project", "setup project", "generate CLAUDE.md", "configure for dotnet-claude-kit"

How

Step 1: Detect or Ask Project Type

Analyze the current directory to determine if this is an existing or greenfield project:

→ Look for .slnx / .sln files
→ If found, scan .csproj files for SDK type:
  - Microsoft.NET.Sdk.Web → web-api or blazor-app
  - Microsoft.NET.Sdk.Worker → worker-service
  - Microsoft.NET.Sdk → class-library
  - Multiple projects with inter-module references → modular-monolith
  - If ambiguous, ask the user

→ If NO solution/project found (greenfield):
  - Ask: "What are you building?"
    - REST API / microservice → web-api
    - Blazor application → blazor-app
    - Background worker / queue processor → worker-service
    - NuGet package / shared library → class-library
    - Multi-module system → modular-monolith
  - Ask: "Project name?"
  - Scaffold the solution structure:
    - dotnet new sln -n ProjectName --format slnx   (modern XML solution format)
    - dotnet new webapi / worker / classlib as appropriate
    - Set up Directory.Build.props with .NET 10 defaults
    - Create src/ and tests/ folder structure

Step 2: Architecture Questionnaire

Load the architecture-advisor skill and ask targeted questions:

  1. Domain complexity — CRUD-heavy, moderate business rules, or rich domain?
  2. Team size — Solo, small team, or large team?
  3. Module boundaries — Single deployable or multiple bounded contexts?
  4. Existing patterns — (existing projects only) Detect conventions via convention-learner skill

→ Recommend: VSA, Clean Architecture, DDD, or Modular Monolith with rationale.

Step 3: Tech Stack Selection

Ask about specific technology choices:

  • Database: PostgreSQL, SQL Server, SQLite, or none yet
  • Auth: JWT, OIDC, Cookie, or none yet
  • Caching: HybridCache, Redis, or none yet
  • Messaging: Wolverine, MassTransit, or none yet
  • Testing: Confirm xUnit + Testcontainers defaults

Step 4: Generate CLAUDE.md

Load the appropriate template from templates/ as a base, then customize:

  • Replace [ProjectName] with the actual project/solution name
  • Set the chosen architecture
  • Configure the tech stack section
  • For existing projects: add project-specific conventions detected in Step 2
  • For greenfield: add recommended conventions based on architecture choice

Before writing, check for an existing CLAUDE.md. Never overwrite it silently.

→ If ./CLAUDE.md does NOT exist:
  - Write the generated content to ./CLAUDE.md

→ If ./CLAUDE.md ALREADY exists:
  - Read it and identify hand-written / custom sections
    (anything not produced by a kit template)
  - Ask: "You already have a CLAUDE.md. How should I proceed?"
    - Merge (recommended) → keep the user's custom sections, add or refresh the
      kit's architecture, tech-stack, and convention sections in place
    - Overwrite → replace it entirely with the freshly generated file
    - Keep → leave the file untouched and print the generated content so the
      user can copy what they want by hand
  - For Merge and Overwrite: back the original up to ./CLAUDE.md.bak first.
    Never discard user content without a .bak backup.

Step 5: Verify Setup

→ Confirm .mcp.json exists or create it (for Roslyn MCP server)
→ For existing projects:
  - Verify the solution file is discoverable
  - Run a quick health check to establish baseline
  - Report: "Project initialized. Run /health-check to see your baseline."
→ For greenfield projects:
  - Verify dotnet build succeeds on the scaffolded solution
  - Report: "Project created. Run /scaffold to add your first feature."

Skills invoked: project-setup, architecture-advisor, convention-learner Agent: dotnet-architect

Example — Existing Project

User: /dotnet-init

Claude: Detected:
  - Solution: MyApp.slnx (3 projects)
  - SDK: Microsoft.NET.Sdk.Web → Web API
  - Database: PostgreSQL (from connection string)
  - Auth: JWT (from appsettings)

  1. Domain complexity? [CRUD-heavy / Moderate / Rich domain]
  > Moderate

  2. Team size? [Solo / Small / Large]
  > Small

  Recommendation: Vertical Slice Architecture
  Generated: ./CLAUDE.md (MyApp + VSA + PostgreSQL + JWT)
  Run /health-check to see your baseline.

Example — Greenfield Project

User: /dotnet-init

Claude: No .sln or .csproj found. Let's set up a new project.

  1. What are you building? [REST API / Blazor / Worker / Library / Modular Monolith]
  > REST API

  2. Project name?
  > OrderService

  3. Domain complexity? [CRUD-heavy / Moderate / Rich domain]
  > CRUD-heavy

  Recommendation: Vertical Slice Architecture

  Creating solution...
  → dotnet new sln -n OrderService --format slnx
  → dotnet new webapi -n OrderService.Api -o src/OrderService.Api
  → dotnet new xunit3 -n OrderService.Tests -o tests/OrderService.Tests
  → Directory.Build.props with .NET 10 defaults
  → Generated: ./CLAUDE.md (OrderService + VSA + scaffold-ready)

  Run /scaffold to add your first feature.

Example — Existing Project With a CLAUDE.md

User: /dotnet-init

Claude: Detected:
  - Solution: MyApp.slnx (3 projects) → Web API
  - Found an existing CLAUDE.md (custom "Deployment" and "Team conventions" sections)

  You already have a CLAUDE.md. How should I proceed?
  [Merge (recommended) / Overwrite / Keep]
  > Merge

  → Backed up original to ./CLAUDE.md.bak
  → Kept your Deployment + Team conventions sections
  → Refreshed architecture (VSA), tech-stack (PostgreSQL + JWT), and convention sections
  Generated: ./CLAUDE.md
  Run /health-check to see your baseline.

Related

  • /plan — Plan before building features
  • /health-check — Assess project health after init
  • /scaffold — Scaffold features using the chosen architecture

Embed badges

Add these to your README to show the skill's verification status.

SkillSafe verified badge
Verified badge
[![SkillSafe verified badge](https://api.skillsafe.ai/v1/badge/@codewithmukesh/dotnet-init/verified)](https://skillsafe.ai/skill/@codewithmukesh/dotnet-init/)
Installs badge
Installs badge
[![Installs badge](https://api.skillsafe.ai/v1/badge/@codewithmukesh/dotnet-init/installs)](https://skillsafe.ai/skill/@codewithmukesh/dotnet-init/)
Scan badge
Scan badge
[![Scan badge](https://api.skillsafe.ai/v1/badge/@codewithmukesh/dotnet-init/scan)](https://skillsafe.ai/skill/@codewithmukesh/dotnet-init/)
Eval pass rate badge
Eval pass rate
[![Eval pass rate badge](https://api.skillsafe.ai/v1/badge/@codewithmukesh/dotnet-init/eval)](https://skillsafe.ai/skill/@codewithmukesh/dotnet-init/)