- Home
- Skills
- Technical Documentation
- README Generator: Professional, Verified READMEs
Works with the AI Agents you already use
README Generator: Professional, Verified READMEs
- **Verifies the quick start**: install, build, the main example and tests are actually run.
Free
README Generator: Professional, Verified READMEs
Example session with this skill installed
Write a proper README for this repo. It's going public next week.
- Read your context and instructions
- Compiled the readme generator
Example output on a sample Node.js library.
Summary
Detected a TypeScript library (orderflow, published on npm) with a CLI, Redis-backed queue, MIT license and GitHub Actions CI. Ran 5 commands, all passing. Wrote a 140-line README.md; kept your existing "Why we built this" section.
| Check | Result |
|---|---|
npm install orderflow | ✅ |
npx orderflow init | ✅ creates orderflow.config.ts |
npx orderflow run --dry | ✅ |
npm run build | ✅ |
npm test | ✅ 86 passed |
README.md (excerpt)
# orderflow
> Lightweight order pipeline for Node.js: queue, retry and audit every order in one package.
[](...) [](...) [](LICENSE)
## Features
- Exactly-once processing with idempotency keys
- Automatic retries with exponential backoff
- Full audit trail per order, queryable from the CLI
## Quick start
npm install orderflow
npx orderflow init
npx orderflow run --dry # processes 3 sample orders
## Configuration
| Variable | Default | Description |
|---|---|---|
| OF_REDIS_URL | redis://localhost:6379 | Queue backend |
| OF_MAX_RETRIES | 5 | Retries per job before dead-letter |
| OF_LOG_LEVEL | info | debug, info, warn, error |
Please add
- A 10-second terminal GIF of
orderflow run(placeholder left under the title) - Roadmap items: not determinable from the code
Next step: review the tagline and features, then commit README.md.
Connects securely to your tools. The creator never sees your data.
About this skill
Your README is the landing page of your project. Most are either empty, outdated or full of commands that no longer work, so visitors leave before trying it.
This skill writes the README a great open-source maintainer would: it sells the project in 10 seconds and gets a newcomer running in 5 minutes, with facts taken from the code, not invented.
What it does
- Understands the project: language, framework, install method (npm, pip, brew, Docker, binary), entry points (CLI, API, server), env vars, license, CI and published package name.
- Verifies the quick start: install, build, the main example and tests are actually run. Only working commands are documented, with real prerequisites (runtime versions from engines, .nvmrc, python_requires).
- Writes real examples: CLI flags, API calls and code snippets taken from your code, with expected output. No foo/bar.
- Builds the configuration table from .env.example,
--helpoutput or the config schema. Never invents options. - Adds structure that converts: one-line value proposition, badges that resolve, benefit-led features, "How it works" (with a Mermaid diagram for non-trivial projects), development, contributing and license.
- Upgrades, doesn't erase: keeps accurate human-written content (story, screenshots, credits) from an existing README.
Adapts to the project type
Library, CLI tool, web app, REST API, template or monorepo. Empty sections are omitted.
Best for
- Preparing an open-source release or Product Hunt / Hacker News launch
- Internal repos nobody can set up
- Portfolio projects that need to look professional
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Technical Documentation.
- Commands that need secrets, paid services or cloud resources are documented but marked as not verified.
- Screenshots and demo GIFs are not generated; it leaves clear placeholders and tells you what to capture.
- Audience, positioning and roadmap may need your input when they cannot be inferred from the code.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
This skill has been used 10 times.
Trust & safety
Security scan
Security scan passed 11 days ago
- Free with an Agensi account
- Software
GitProject toolchain
Uses your project's own toolchain, such as Node.js with npm or Python with pip.
Permissions
Read Files: reads code, manifests, configs, CI and existing docs. Terminal: runs your install, build, example and test commands to verify them. Write Files: creates or updates README.md only. Network: only what your own install commands need (e.g. npm install).