Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
# Mission: [Project Name]> [One-line pitch describing the project's core purpose]## Product Pitch
[2-3 paragraph description of what the product does, why it exists, and what problem it solves]
## Target Users### Primary Users-**[User Type 1]**: [Description and needs]
-**[User Type 2]**: [Description and needs]
: [Description and needs]
: [Description of the problem]
: [Description of the problem]
: [Description of the problem]
: [How this product solves the problem]
: [How this product solves the problem]
: [How this product solves the problem]
| Metric | Current | Target | Timeframe |
|--------|---------|--------|-----------|
| [Metric 1] | [Current value] | [Target value] | [When] |
| [Metric 2] | [Current value] | [Target value] | [When] |
| [Metric 3] | [Current value] | [Target value] | [When] |
: [Description]
: [Description]
: [Description]
: [How we differ]
: [How we differ]
Things explicitly out of scope:
[Non-goal 1]
[Non-goal 2]
[Non-goal 3]
---
### Secondary Users
-
**[User Type 3]**
## Pain Points Addressed
### Before This Product
1.
**[Pain Point 1]**
2.
**[Pain Point 2]**
3.
**[Pain Point 3]**
### After This Product
1.
**[Solution 1]**
2.
**[Solution 2]**
3.
**[Solution 3]**
## Success Metrics
## Differentiators
### What Makes This Unique
1.
**[Differentiator 1]**
2.
**[Differentiator 2]**
3.
**[Differentiator 3]**
### Competitive Landscape
-
**[Competitor 1]**
-
**[Competitor 2]**
## Non-Goals
-
-
-
*Last Updated: [Date]*
*Version: 1.0.0*
2. tech-stack.md
# Tech Stack: [Project Name]> Technical architecture and technology choices## Overview
| Category | Technology | Version | Purpose |
|----------|-----------|---------|---------|
| Language | Python | 3.11+ | Primary development |
| Package Manager | UV | Latest | Fast dependency management |
| Testing | pytest | 7.4+ | Test framework |
| Visualization | Plotly | 5.15+ | Interactive charts |
| Data | Pandas | 2.0+ | Data processing |
## Core Technologies### Python 3.11+**Why**: Modern async support, performance improvements, type hints
**Usage**: All source code in `src/`### UV Package Manager**Why**: 10-100x faster than pip, reliable lockfiles
**Usage**: `uv venv`, `uv pip install`### pytest**Why**: Industry standard, excellent fixtures, plugins
**Usage**: All tests in `tests/`### Plotly**Why**: Interactive plots, HTML export, professional appearance
**Usage**: All visualizations must be interactive (no static matplotlib)
### Pandas**Why**: Data manipulation, time series, CSV handling
**Usage**: Data loading and transformation
## Development Tools
| Tool | Purpose | Configuration |
|------|---------|---------------|
| ruff | Linting | pyproject.toml |
| black | Formatting | pyproject.toml |
| mypy | Type checking | pyproject.toml |
| pytest-cov | Coverage | pytest.ini |
## Infrastructure### Version Control-**Git**: Source control
-**GitHub**: Remote repository
-**Branch Strategy**: main → feature branches → PR
### CI/CD-**GitHub Actions**: Automated testing
-**Coverage**: Minimum 80%
## Data Storage
| Type | Location | Format |
|------|----------|--------|
| Raw data | data/raw/ | CSV, JSON |
| Processed | data/processed/ | CSV, Parquet |
| Results | data/results/ | CSV, JSON |
| Reports | reports/ | HTML |
## External Dependencies### APIs- [API 1]: [Purpose]
- [API 2]: [Purpose]
### Services- [Service 1]: [Purpose]
- [Service 2]: [Purpose]
## Decision Rationale### Why Python?- Strong ecosystem for data analysis
- Excellent library support (Pandas, NumPy, Plotly)
- Team expertise
- Integration with existing tools
### Why UV over pip?- Significantly faster installation
- Reliable dependency resolution
- Lockfile support
- workspace-hub standard
### Why Plotly over Matplotlib?- Interactive by default
- Better HTML export
- Modern API
- workspace-hub HTML reporting standard
---
*Last Updated: [Date]**Version: 1.0.0*
3. roadmap.md
# Roadmap: [Project Name]> Development phases and milestones## Vision
[Long-term vision for the product - where it will be in 1-2 years]
## Current Phase**Phase [N]: [Phase Name]**- Status: [In Progress / Planning / Complete]
- Target: [Date]
- Progress: [X]%
## Phase Overview
# Decision Log: [Project Name]> Record of architectural and design decisions## How to Use This Document
Document significant technical decisions using the format below. Include context, options considered, and rationale.
## Decision Template```markdown
### DEC-XXX: [Decision Title]
**Date**: YYYY-MM-DD
**Status**: [Proposed | Accepted | Deprecated | Superseded]
**Deciders**: [Names or roles]
#### Context
[What is the issue or opportunity?]
#### Options Considered
1. **Option A**: [Description]
- Pros: [Benefits]
- Cons: [Drawbacks]
2. **Option B**: [Description]
- Pros: [Benefits]
- Cons: [Drawbacks]
#### Decision
[Which option was chosen and why]
#### Consequences
- Positive: [Good outcomes]
- Negative: [Trade-offs accepted]
#### Related
- [Links to related decisions, issues, docs]
Decisions
DEC-001: Package Manager Selection
Date: 2026-01-01
Status: Accepted
Deciders: Engineering Team
Context
Need to select a Python package manager for dependency management across the project.
Options Considered
pip + requirements.txt
Pros: Universal, simple
Cons: Slow, no lockfile
poetry
Pros: Modern, lockfile support
Cons: Slower than UV
UV
Pros: Very fast, lockfiles, drop-in pip replacement