project-init
Auto-generate project-level CLAUDE.md from templates based on detected stack.
Skill by Leo-Atienza
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/leo-atienza-project-init
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/project-init
What's inside
- Skill name
project-init- Description
- Auto-generate project-level CLAUDE.md from templates based on detected stack. Adapted from Ruflo's 30+ template system. Invoke from /new after project directory is set up.
- License
- MIT
- Instructions
- 241 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Project Init — CLAUDE.md Template Generator
Generates a project-level CLAUDE.md tailored to the detected tech stack.
Eliminates boilerplate setup and ensures consistent project conventions.
## Detection Logic
Scan the project root for stack indicators:
```
package.json + next.config.* → Next.js template
package.json + expo → Expo/React Native template
package.json (generic) → Node.js template
pubspec.yaml → Flutter/Dart template
requirements.txt | pyproject.toml → Python template
Cargo.toml → Rust template
go.mod → Go template
Gemfile → Ruby/Rails template
*.sln | *.csproj → .NET template
build.gradle* → Java/Kotlin template
Dockerfile (no app code) → DevOps template
No indicators → Generic template
```
If multiple indicators exist, use the primary one (first match in order above).
## Template: Next.js
```markdown
# Project Instructions
## Stack
- Next.js (App Router), React, TypeScript
- Tailwind CSS for styling
## Conventions
- Use Server Components by default, 'use client' only when needed
- File conventions: page.tsx, layout.tsx, loading.tsx, error.tsx
- Data fetching in Server Components (no useEffect for initial data)
- Route handlers in app/api/ with proper HTTP status codes
## Testing
- Jest + React Testing Library for unit tests
- Playwright for E2E (if configured)
- Run: `npm test` (unit), `npx playwright test` (E2E)
## Commands
- Dev: `npm run dev`
- Build: `npm run build`
- Lint: `npm run lint`
- Format: `npx prettier --write .`
```
## Template: Python
```markdown
# Project Instructions
## Stack
- Python 3.11+
- Package manager: uv (preferred) or pip
## Conventions
- Type hints on all function signatures
- snake_case everywhere, UPPER_SNAKE for constants
- Async/await for I/O-bound operations
- Pydantic for data validation
## Testing
- pytest with fixtures
- Run: `pytest` or `uv run pytest`
- Coverage: `pytest --cov`
## Quality
- Formatter: ruff format
- Linter: ruff check
- Type checker: ty or mypy
```
## Template: Flutter/Dart
```markdown
# Project Instructions
## Stack
- Flutter, Dart
- State management: [detect from pubspec.yaml — riverpod/bloc/provider]
## Conventions
- Widgets: small, composable, single-purpose
- File structure: lib/features/{name}/ with models, views, controllers
- const constructors wherever possible
- Named parameters for widget constructors
## Testing
- flutter test for unit/widget tests
- integration_test/ for integration tests
- Run: `flutter test`
## Commands
- Dev: `flutter run`
- Build: `flutter build apk` / `flutter build ios`
- Format: `dart format .`
- Analyze: `dart analyze`
```
## Template: Go
```markdown
# Project Instructions
## Stack
- Go 1.22+
## Conventions
- Error handling: always check returned errors, wrap with fmt.Errorf
- Interfaces: accept interfaces, return structs
- Package names: short, lowercase, no underscores
- Context: pass as first parameterFrom Leo-Atienza/atlas-claude, MIT.
More from Leo-Atienza/atlas-claude
- 0.519score
self-evolve
Leo-Atienza
Detect capability gaps and autonomously create skills, add free MCP servers, or organize agent teams.
SkillCFreshness: C★ 4Claude Code +9 - 0.519score
design-audit
Leo-Atienza
Run technical quality checks across accessibility, performance, theming, responsive design, and anti-patterns.
SkillCFreshness: C★ 4Claude Code +9 - 0.519score
codebase-knowledge-graph
Leo-Atienza
Router skill for graph-based codebase/corpus queries.
SkillCFreshness: C★ 4Claude Code +9 - 0.519score
smart-swarm
Leo-Atienza
Automatically organize multi-agent teams for complex tasks.
SkillCFreshness: C★ 4Claude Code +9 - 0.513score
weekly-cleanup-scan
Leo-Atienza
Weekly cleanup scan — hooks, disk usage, stale files, MCP health
SkillCFreshness: C★ 4Claude Code +9 - 0.513score
monthly-evolution-report
Leo-Atienza
Monthly evolution report — knowledge growth, usage analysis, optimization suggestions
SkillCFreshness: C★ 4Claude Code +9 - 0.513score
weekly-memory-maintenance
Leo-Atienza
Weekly memory maintenance — prune sessions >30, check INDEX/topics consistency, evolution health
SkillCFreshness: C★ 4Claude Code +9 - 0.513score
skill-usage-audit
Leo-Atienza
Monthly audit of archived skills — auto-unarchive used skills, auto-remove unused ones
SkillCFreshness: C★ 4Claude Code +9 - 0.508score
weekly-dream
Leo-Atienza
Weekly memory consolidation — runs /dream to orient, gather signal, merge, and prune memory files
SkillCFreshness: C★ 4Claude Code +9 - 0.494score
brand-guidelines
Leo-Atienza
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel.
SkillCFreshness: C★ 4Claude Code +9 - 0.280score
css-first-ui
Leo-Atienza
No description from the publisher yet.
SkillCFreshness: C★ 4Claude Code +9 - 0.280score
ai-native-ui
Leo-Atienza
No description from the publisher yet.
SkillCFreshness: C★ 4Claude Code +9
Signals
- Stars
- 4
- Forks
- 5
- Open issues
- 0
- Last commit (this folder)
- Jul 20, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/Leo-Atienza/atlas-claude
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Leo-Atienza
- 0.519score
design-audit
Leo-Atienza
Run technical quality checks across accessibility, performance, theming, responsive design, and anti-patterns.
SkillCFreshness: C★ 4Claude Code +9 - 0.519score
self-evolve
Leo-Atienza
Detect capability gaps and autonomously create skills, add free MCP servers, or organize agent teams.
SkillCFreshness: C★ 4Claude Code +9 - 0.519score
smart-swarm
Leo-Atienza
Automatically organize multi-agent teams for complex tasks.
SkillCFreshness: C★ 4Claude Code +9 - 0.519score
codebase-knowledge-graph
Leo-Atienza
Router skill for graph-based codebase/corpus queries.
SkillCFreshness: C★ 4Claude Code +9 - 0.513score
weekly-memory-maintenance
Leo-Atienza
Weekly memory maintenance — prune sessions >30, check INDEX/topics consistency, evolution health
SkillCFreshness: C★ 4Claude Code +9 - 0.513score
monthly-evolution-report
Leo-Atienza
Monthly evolution report — knowledge growth, usage analysis, optimization suggestions
SkillCFreshness: C★ 4Claude Code +9
Cite this page
project-init skill. Agent Mart. https://agentmart.dev/skills/leo-atienza-project-init (last updated 2026-10-24)