claude-code-starter 0.1.2 → 0.1.3

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -1,6 +1,6 @@
1
1
  # Claude Code Starter
2
2
 
3
- A lightweight starter kit for AI-assisted development with Claude Code.
3
+ Intelligent CLI that bootstraps Claude Code configurations tailored to your project's tech stack.
4
4
 
5
5
  ## Quick Start
6
6
 
@@ -12,27 +12,251 @@ claude
12
12
 
13
13
  ## What It Does
14
14
 
15
- Sets up your project with:
16
- - `.claude/CLAUDE.md` - Instructions for Claude
17
- - `.claude/commands/` - Slash commands (`/task`, `/status`, `/done`, `/analyze`)
18
- - `.claude/skills/` - Methodology guides (debugging, testing, patterns)
19
- - `.claude/state/task.md` - Task tracking
15
+ 1. **Analyzes your repository** - Detects languages, frameworks, tools, and patterns
16
+ 2. **Generates tailored configurations** - Creates skills, agents, and rules for your stack
17
+ 3. **Sets up development workflow** - Task tracking, commands, and methodology guides
18
+
19
+ ## Tech Stack Detection
20
+
21
+ Automatically detects and configures for:
22
+
23
+ | Category | Detected |
24
+ | -------------- | ------------------------------------------------------------------- |
25
+ | **Languages** | TypeScript, JavaScript, Python, Go, Rust, Java, Ruby, and more |
26
+ | **Frameworks** | Next.js, React, Vue, Svelte, FastAPI, Django, NestJS, Express, etc. |
27
+ | **Tools** | npm, yarn, pnpm, bun, pip, cargo, go modules |
28
+ | **Testing** | Jest, Vitest, Pytest, Go test, Rust test |
29
+ | **Linting** | ESLint, Biome, Ruff, Pylint |
30
+
31
+ ## Generated Configurations
32
+
33
+ Based on your stack, creates:
34
+
35
+ - **📚 Skills** - Framework-specific patterns (e.g., Next.js App Router, FastAPI endpoints)
36
+ - **🤖 Agents** - Specialized assistants (code reviewer, test writer)
37
+ - **📏 Rules** - Language conventions (TypeScript strict mode, Python PEP 8)
38
+ - **⚡ Commands** - Workflow shortcuts (`/task`, `/status`, `/done`, `/analyze`)
20
39
 
21
40
  ## Commands
22
41
 
23
- | Command | Description |
24
- |---------|-------------|
25
- | `/task <desc>` | Start a new task |
26
- | `/status` | Show current task |
27
- | `/done` | Mark task complete |
42
+ | Command | Description |
43
+ | ----------------- | ------------------- |
44
+ | `/task <desc>` | Start a new task |
45
+ | `/status` | Show current task |
46
+ | `/done` | Mark task complete |
28
47
  | `/analyze <area>` | Deep dive into code |
29
48
 
30
- ## Options
49
+ ## CLI Options
31
50
 
32
51
  ```bash
33
- npx claude-code-starter --help # Show help
34
- npx claude-code-starter --force # Overwrite existing files
52
+ npx claude-code-starter # Interactive mode
53
+ npx claude-code-starter -y # Non-interactive (use defaults)
54
+ npx claude-code-starter -f # Force overwrite existing files
55
+ npx claude-code-starter -V # Verbose output
56
+ npx claude-code-starter --help # Show help
57
+ ```
58
+
59
+ ## Example Output
60
+
61
+ ```
62
+ ╔═══════════════════════════════════╗
63
+ ║ Claude Code Starter ║
64
+ ║ AI-Assisted Development Setup ║
65
+ ╚═══════════════════════════════════╝
66
+
67
+ 🔍 Analyzing repository...
68
+
69
+ 📊 Tech Stack Analysis
70
+ Language: TypeScript
71
+ Framework: Next.js
72
+ Package Manager: bun
73
+ Testing: vitest
74
+
75
+ ✅ Configuration complete! (14 files)
76
+
77
+ Generated for your stack:
78
+ 📚 4 skills (pattern-discovery, testing-methodology, nextjs-patterns, ...)
79
+ 🤖 2 agents (code-reviewer, test-writer)
80
+ 📏 2 rules
81
+ ```
82
+
83
+ ## Project Structure
84
+
85
+ After running, your project will have:
86
+
87
+ ```
88
+ .claude/
89
+ ├── CLAUDE.md # Project-specific instructions
90
+ ├── settings.json # Permissions configuration
91
+ ├── agents/ # Specialized AI agents
92
+ │ ├── code-reviewer.md
93
+ │ └── test-writer.md
94
+ ├── commands/ # Slash commands
95
+ │ ├── task.md
96
+ │ ├── status.md
97
+ │ ├── done.md
98
+ │ └── analyze.md
99
+ ├── rules/ # Code style rules
100
+ │ ├── typescript.md # (or python.md, etc.)
101
+ │ └── code-style.md
102
+ ├── skills/ # Framework-specific patterns
103
+ │ ├── pattern-discovery.md
104
+ │ ├── systematic-debugging.md
105
+ │ ├── testing-methodology.md
106
+ │ └── nextjs-patterns.md # (or fastapi-patterns.md, etc.)
107
+ └── state/
108
+ └── task.md # Current task tracking
109
+ ```
110
+
111
+ ## How It Works
112
+
113
+ ### Architecture
114
+
115
+ ```mermaid
116
+ flowchart LR
117
+ A[CLI] --> B[Analyzer]
118
+ B --> C[Generator]
119
+ C --> D[.claude/]
120
+
121
+ B -->|reads| E[package.json]
122
+ B -->|reads| F[config files]
123
+ B -->|reads| G[lock files]
124
+
125
+ C -->|creates| H[skills/]
126
+ C -->|creates| I[agents/]
127
+ C -->|creates| J[rules/]
128
+ C -->|creates| K[commands/]
129
+ ```
130
+
131
+ ### Artifact Generation
132
+
133
+ All artifacts are **generated dynamically** based on your detected tech stack. There are no static templates - content is created specifically for your project.
134
+
135
+ | Artifact Type | Generation Logic |
136
+ |---------------|------------------|
137
+ | **CLAUDE.md** | Generated with project name, description, and detected stack |
138
+ | **settings.json** | Generated with safe default permissions |
139
+ | **Skills** | Core skills + framework-specific patterns (if detected) |
140
+ | **Agents** | Code reviewer and test writer agents |
141
+ | **Rules** | Language-specific conventions + general code style |
142
+ | **Commands** | Task workflow commands (/task, /status, /done, /analyze) |
143
+
144
+ ### Conflict Resolution
145
+
146
+ When running on an existing project with `.claude/` configuration:
147
+
148
+ | Scenario | Behavior |
149
+ |----------|----------|
150
+ | **New file** | Created |
151
+ | **Existing file** | Skipped (preserved) |
152
+ | **With `-f` flag** | Overwritten |
153
+ | **state/task.md** | Always preserved |
154
+
155
+ ### Detection Sources
156
+
157
+ The CLI examines these files to detect your stack:
158
+
159
+ | Source | Detects |
160
+ |--------|---------|
161
+ | `package.json` | JS/TS frameworks, testing, linting |
162
+ | `pyproject.toml` / `requirements.txt` | Python frameworks, tools |
163
+ | `go.mod` | Go and its web frameworks |
164
+ | `Cargo.toml` | Rust and its frameworks |
165
+ | `*.lockfile` | Package manager (bun.lockb, yarn.lock, etc.) |
166
+ | `.github/workflows/` | CI/CD platform |
167
+ | Config files | Linters, formatters, bundlers |
168
+
169
+ ### Framework-Specific Patterns
170
+
171
+ When a framework is detected, additional skills are generated:
172
+
173
+ | Framework | Generated Skill |
174
+ |-----------|-----------------|
175
+ | Next.js | `nextjs-patterns.md` - App Router, Server Components |
176
+ | React | `react-components.md` - Hooks, component patterns |
177
+ | FastAPI | `fastapi-patterns.md` - Async endpoints, Pydantic |
178
+ | NestJS | `nestjs-patterns.md` - Modules, decorators, DI |
179
+
180
+ ### Artifact Examples
181
+
182
+ <details>
183
+ <summary><b>📚 Skill Example</b> (pattern-discovery.md)</summary>
184
+
185
+ ```markdown
186
+ ---
187
+ description: "Systematic approach to discovering patterns in unfamiliar codebases"
188
+ ---
189
+
190
+ # Pattern Discovery
191
+
192
+ ## When to Use
193
+ - Starting work on unfamiliar code
194
+ - Investigating "how does X work here?"
195
+ - Before making changes to existing systems
196
+
197
+ ## Approach
198
+ 1. Find entry points (main, index, app)
199
+ 2. Trace data flow through the system
200
+ 3. Identify naming conventions
201
+ 4. Map directory structure to responsibilities
202
+ ```
203
+ </details>
204
+
205
+ <details>
206
+ <summary><b>🤖 Agent Example</b> (code-reviewer.md)</summary>
207
+
208
+ ```markdown
209
+ ---
210
+ name: "Code Reviewer"
211
+ description: "Reviews code for quality, patterns, and potential issues"
212
+ model: "sonnet"
213
+ tools: ["Read", "Glob", "Grep"]
214
+ ---
215
+
216
+ # Code Reviewer Agent
217
+
218
+ You are a code reviewer focused on:
219
+ - Code quality and readability
220
+ - Potential bugs and edge cases
221
+ - Performance considerations
222
+ - Security vulnerabilities
35
223
  ```
224
+ </details>
225
+
226
+ <details>
227
+ <summary><b>📏 Rule Example</b> (typescript.md)</summary>
228
+
229
+ ```markdown
230
+ ---
231
+ paths: ["**/*.ts", "**/*.tsx"]
232
+ ---
233
+
234
+ # TypeScript Conventions
235
+
236
+ - Use strict mode (`"strict": true`)
237
+ - Prefer `interface` over `type` for object shapes
238
+ - Use explicit return types on exported functions
239
+ - Avoid `any`, prefer `unknown` for truly unknown types
240
+ ```
241
+ </details>
242
+
243
+ <details>
244
+ <summary><b>⚡ Command Example</b> (task.md)</summary>
245
+
246
+ ```markdown
247
+ ---
248
+ description: "Start or switch to a new task"
249
+ ---
250
+
251
+ # /task
252
+
253
+ Update `.claude/state/task.md` with:
254
+ - Task description from user input
255
+ - Status: In Progress
256
+ - Empty decisions/notes sections
257
+ - Clear next steps
258
+ ```
259
+ </details>
36
260
 
37
261
  ## Tip
38
262
 
@@ -43,6 +267,22 @@ echo ".claude/" >> ~/.gitignore
43
267
  git config --global core.excludesfile ~/.gitignore
44
268
  ```
45
269
 
270
+ ## Development
271
+
272
+ ```bash
273
+ # Install dependencies
274
+ bun install
275
+
276
+ # Run in development mode
277
+ bun run dev
278
+
279
+ # Run tests
280
+ bun test
281
+
282
+ # Build
283
+ bun run build
284
+ ```
285
+
46
286
  ## License
47
287
 
48
288
  MIT