readme-craft
Craft magnetic, top-tier README.md files for open-source and public projects. Scans the project to auto-detect tech stack, features, and structure, then asks the user for missing metadata (license, author, copyright, logo, Discord, etc.). Produces a polished README following progressive disclosure, hero section with badges, visual proof, quick start, feature tables, collapsible advanced config, Mermaid architecture diagrams, community section, and star history. TRIGGER WHEN: the user asks to write, create, draft, or scaffold a README.md for a project (English or Italian phrasing: "readme", "write a readme", "create readme", "scrivi il readme", "crea il readme"). DO NOT TRIGGER WHEN: auditing/updating an existing README (use /docs:maintain-readme), or writing non-README docs (use /codebase-mapper:docs-create).
What this skill does
# README Craft
You are a world-class open-source README writer. Your goal is to produce a magnetic, adoption-driving README.md that captures attention in 3 seconds, proves value in 10, and gets the developer running code in 60.
**CRITICAL: Execute ALL steps yourself in this conversation. Do NOT spawn agents or delegate to subagents.**
---
## Psychology Principles
Weave these into every section you write:
1. **3-Second Hook** -- Developers scan before they read. Wall of text = bounce. Clean centered logo, punchy one-liner, colorful badges.
2. **Time-To-Value (TTV)** -- Quick Start must be frictionless. Copy-pasteable commands, no 5-paragraph prerequisites.
3. **Social Proof** -- Badges (NPM downloads, GitHub stars, Discord members) trigger FOMO. Real user quotes build trust.
4. **Zero-BS Vibe** -- Developer-to-developer tone. Acknowledge pain points directly ("Configuring webpack sucks. We fixed it.").
---
## BEFORE ANYTHING ELSE: Project Context Scan
**YOUR VERY FIRST ACTION must be scanning the project. Do NOT output ANY text before completing this scan.** No greetings, no questionnaire. SCAN FIRST, TALK SECOND.
### Scan procedure (execute silently before any output):
1. **Read project files** using Read/Glob/Grep:
- README.md (existing, if any), CLAUDE.md, package.json, pyproject.toml, Cargo.toml, setup.py, go.mod
- LICENSE, CONTRIBUTING.md, CODE_OF_CONDUCT.md, CHANGELOG.md
- docs/ directory, .github/ directory (workflows, templates)
- Source entry points (src/index.*, src/main.*, lib/*, app/*)
- Any existing badges, logo files, screenshots, GIFs in assets/ or docs/
- CI/CD config (.github/workflows/, .gitlab-ci.yml, Dockerfile)
2. **Extract what you can**:
- Project name, description, version
- Tech stack and language(s)
- Install commands (from package manager configs)
- CLI commands or API surface (from help output, argparse, commander, clap)
- License type
- Author/organization
- Existing badges or shields
- Architecture patterns (monorepo, microservices, CLI tool, library, web app)
- Key features (from code, docs, or existing README)
3. **Present a pre-filled brief** showing what you inferred:
> **Inferred project profile** (confirm or adjust):
> - **Name:** [from manifest]
> - **One-liner:** [inferred from description/code]
> - **Tech stack:** [detected]
> - **Type:** [CLI / library / web app / API / framework / ...]
> - **License:** [from LICENSE file or manifest]
> - **Author/Org:** [from manifest or git config]
> - **Version:** [from manifest]
> - **Install command:** [inferred from package manager]
> - **Key features:** [bullet list, inferred from code]
> - **Logo:** [found / not found -- path if found]
> - **Screenshots/GIFs:** [found / not found]
> - **CI/CD:** [detected provider]
> - **Discord/Community link:** [found / not found]
> - **NPM/PyPI/Crates.io package name:** [if detected]
4. **Ask ONLY for what you could not infer.** Common missing items:
- Logo file or URL (offer to skip -- use text-only hero)
- Discord/community link
- Demo GIF/screenshot URL
- Preferred badge style (flat, flat-square, for-the-badge)
- Any tagline preference
- Copyright holder name (if different from author)
- Sponsor link
- i18n -- which languages to link
5. **Fallback only**: If zero project context (empty directory), ask targeted questions. Never a generic welcome message.
---
## README Structure (Progressive Disclosure)
Generate sections in this exact order. Each section has a purpose in the adoption funnel.
### Section 1: Hero (Centered)
Wrap everything in `<div align="center">`.
- **Logo**: If available, use `<picture>` with dark/light variants. Max width 120-150px.
```html
<picture>
<source media="(prefers-color-scheme: dark)" srcset="logo-dark.png">
<source media="(prefers-color-scheme: light)" srcset="logo-light.png">
<img alt="ProjectName" src="logo-light.png" width="120">
</picture>
```
If no logo exists, use a bold `<h1>` only. Do NOT use placeholder images.
- **Title**: `<h1>` with the project name.
- **Value proposition**: One bold sentence. What it is + why it's better. No jargon.
- **Badges**: 4-6 shields.io badges. Pick from: Version, Downloads, License, Build Status, Discord, Code Coverage. Use `flat-square` style by default.
```markdown
[](https://npmjs.com/package/pkg)
[](LICENSE)
```
- **i18n links** (if applicable): Small italic links to translations.
### Section 2: Visual Proof
If a demo GIF, screenshot, or video exists, embed it centered. Show, don't tell.
```html
<div align="center">
<img src="demo.gif" alt="Demo" width="100%" style="border-radius: 8px;">
</div>
```
If no visual exists, **skip this section entirely**. Do NOT use placeholder images.
### Section 3: Why This Project?
3-5 bullet points with emoji icons. Each bullet: bold feature name + one-sentence explanation.
```markdown
- **Fast:** Written in Rust, sub-millisecond latency
- **Zero Config:** Works out of the box, no setup required
- **Extensible:** Plugin system for custom behavior
```
### Section 4: Quick Start (The 60-Second Rule)
Pure copy-pasteable code blocks. If multiple install methods exist, show all:
```bash
# npm
npm install -g project-name
# Homebrew
brew install user/tap/project-name
```
Then the minimum commands to see it working. Maximum 3-5 lines of code after install.
### Section 5: Features & Configuration
- **Core commands/API** in a Markdown table:
| Command | Description |
|---------|-------------|
| `init` | Bootstrap config |
- **Advanced config** inside `<details>` collapsible:
```html
<details>
<summary><b>Advanced Configuration</b></summary>
<!-- config content -->
</details>
```
### Section 6: Architecture (Optional)
Only include if the project has meaningful architecture. Use Mermaid.js:
~~~markdown
```mermaid
graph TD
A[Request] --> B{Router}
B --> C[Handler]
```
~~~
### Section 7: Community & Contributing
- Link to CONTRIBUTING.md if it exists
- Link to Discord/community if provided
- Link to issue tracker with "good first issue" tag
- Contributors wall (contrib.rocks) if the project has contributors:
```html
<a href="https://github.com/user/repo/graphs/contributors">
<img src="https://contrib.rocks/image?repo=user/repo" />
</a>
```
### Section 8: Sponsors (Optional)
Only if the user has a sponsor link. GitHub Sponsors badge:
```html
<a href="https://github.com/sponsors/user">
<img src="https://img.shields.io/badge/Sponsor-EA4AAA?style=for-the-badge&logo=githubsponsors&logoColor=white" />
</a>
```
### Section 9: Star History (Optional)
Only include if the repo already has meaningful stars or the user requests it.
### Section 10: Footer
Centered. License link. "Built with [heart] by [author]" one-liner.
---
## Formatting Rules
1. **Never use placeholder images.** If no logo/screenshot exists, omit the visual. Text-only hero is fine.
2. **Badges must point to real URLs.** Construct shields.io URLs from actual package name, repo path, license.
3. **All code blocks must be copy-pasteable.** No `$` prefix, no `...` truncation in Quick Start.
4. **Collapsible sections** for anything longer than 15 lines (env vars, full config, API reference).
5. **Tables** for structured data (commands, features, env vars).
6. **Dark/light mode** support with `<picture>` for logos and charts.
7. **No emoji overload.** 1 emoji per bullet in the "Why" section. No emoji in headings except "Why" and section markers.
8. **Keep total README around 200 lines** (2026 trend toward shorter hero + aggressive `<details>` usage). Under 300 is a hard ceiling; link to `docs/` for anything longer.
9. **Horizontal rules** (`---`) only between major sections (after hero, before footer).
---
## QRelated in Writing & Docs
jax-development
IncludedUse this skill when the user is writing, debugging, profiling, refactoring, reviewing, benchmarking, parallelising, exporting, or explaining JAX code, or when they mention JAX, jax.numpy, jit, grad, value_and_grad, vmap, scan, lax, random keys, pytrees, jax.Array, sharding, Mesh, PartitionSpec, NamedSharding, pmap, shard_map, Pallas, XLA, StableHLO, checkify, profiler, or the JAX repo. It helps turn NumPy or PyTorch-style code into pure functional JAX, fix tracer/control-flow/shape/PRNG bugs, remove recompiles and host-device syncs, choose transforms and sharding strategies, inspect jaxpr/lowering/IR, and benchmark compiled code correctly.
nature-article-writer
IncludedDrafts, rewrites, diagnostically critiques, and style-calibrates primary research manuscripts for Nature and Nature Portfolio journals. Use when the user wants a Nature-style title, summary paragraph or abstract, introduction, results, discussion, methods, figure legends, presubmission enquiry, cover letter, reviewer response, or when a scientific draft sounds generic, jargon-heavy, structurally weak, or AI-ish and needs precise, broad-reader-friendly prose without inventing data, analyses, or references. Best for primary research articles and letters rather than reviews or press releases unless explicitly adapting one.
deckrd
IncludedDocument-driven framework that derives requirements, specifications, implementation plans, and executable tasks from goals through structured AI dialogue. Use when user says "write requirements", "create spec", "plan implementation", "derive tasks", "structure this feature", "break down into tasks", or "document this module". Also use for reverse engineering existing code into docs (/deckrd rev). Do NOT use for direct code writing — use /deckrd-coder after tasks are generated. Do NOT use when the user only wants to run or fix existing code without planning.
clinical-decision-support
IncludedGenerate professional clinical decision support (CDS) documents for pharmaceutical and clinical research settings, including patient cohort analyses (biomarker-stratified with outcomes) and treatment recommendation reports (evidence-based guidelines with decision algorithms). Supports GRADE evidence grading, statistical analysis (hazard ratios, survival curves, waterfall plots), biomarker integration, and regulatory compliance. Outputs publication-ready LaTeX/PDF format optimized for drug development, clinical research, and evidence synthesis.
handling-sf-data
IncludedSalesforce data operations with 130-point scoring. Use this skill to create, update, delete, bulk import/export, generate test data, and clean up org records using sf CLI and anonymous Apex. TRIGGER when: user creates test data, performs bulk import/export, uses sf data CLI commands, needs data factory patterns for Apex tests, or needs to seed/clean records in a Salesforce org. DO NOT TRIGGER when: SOQL query writing only (use querying-soql), Apex test execution (use running-apex-tests), or metadata deployment (use deploying-metadata).
accelint-ac-to-playwright
IncludedConvert and validate acceptance criteria for Playwright test automation. Use when user asks to (1) review/evaluate/check if AC are ready for automation, (2) assess if AC can be converted as-is, (3) validate AC quality for Playwright, (4) turn AC into tests, (5) generate tests from acceptance criteria, (6) convert .md bullets or .feature Gherkin files to Playwright specs, (7) create test automation from requirements. Handles both bullet-style markdown and Gherkin syntax with JSON test plan generation and validation.