agentsclimarketplace

Readme writing

Skill easyzoom/aix-skills/skills/readme-writing

Use when creating, rewriting, or improving README files for public repositoriesFrom its SKILL.md

Install
npx -y skills add easyzoom/aix-skills --skill readme-writing

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 24 stars24 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

4.1 KB, 840 tokens by cl100k_base, as published. Nobody here has run it

README Writing

Overview

Write README files that help a visitor understand the project in the first screen, trust it within one minute, and try it without guessing. A good README is a landing page, setup guide, and trust signal at the same time.

When To Use

Use this skill when:

  • Creating a new README.md.
  • Rewriting a README for a public repository.
  • Improving a README's clarity, structure, visuals, or quickstart.
  • Adding badges, screenshots, diagrams, language links, or contribution sections.
  • Making a repository more attractive for GitHub readers.

Do not use it for long-form product docs, API references, or internal-only runbooks unless the user specifically wants a README-style entry page.

Workflow

  1. Identify the primary reader. Decide whether the README is for users, developers, contributors, evaluators, or agents. Put the most important reader first.

  2. Write the first screen. Include project name, one-sentence value proposition, optional language links, visual banner or screenshot, and trust badges.

  3. Explain the project quickly. State what it does, who it is for, and why it is different. Avoid marketing fluff that does not help someone decide whether to continue.

  4. Show the fastest path to value. Provide a short quickstart with commands that can be copied. Include expected output when it reduces uncertainty.

  5. Show structure and capabilities. List key directories, features, or modules. Use tables for scanability when comparing multiple items.

  6. Add proof. Include screenshots, SVG diagrams, badges, test commands, examples, or links to demos. Prefer local assets for long-lived public repos.

  7. Cover contribution and license. Link to contribution docs, issue templates, validation commands, and license.

  8. Verify the README. Check links, commands, image paths, and consistency with the actual repository.

Recommended Structure

# Project Name

Language links, badges, and hero image.

One-sentence value proposition.

## Why This Exists
## Features
## Quickstart
## Examples
## Project Structure
## Contributing
## License

Adjust the sections to the project. Keep the README useful, not ceremonial.

Visual Guidelines

  • Use a hero image, product screenshot, or architecture diagram when it helps visitors understand the project faster.
  • Prefer local assets under docs/assets/ for stability.
  • Use badges sparingly: license, tests, package version, or public readiness are usually enough.
  • Add alt text for images.
  • Avoid huge images that push all useful text below the fold.

Multilingual README Pattern

For multilingual repositories:

  • Keep README.md as the default language most GitHub visitors should see.
  • Add language links near the top.
  • Use files like README.zh-CN.md for full translations.
  • Keep shared sections such as features, quickstart, and skill lists consistent across languages.

Verification

Before claiming completion:

  • Confirm every referenced file path exists.
  • Confirm links point to the right local files or public URLs.
  • Run repository validation or tests if README commands mention them.
  • Check that badges and images render from valid URLs or local files.
  • Re-read the first screen and verify it answers: what is this, why care, how do I start?

Common Failures

  • Starting with a long backstory instead of a clear value proposition.
  • Including installation commands that do not match the repository.
  • Adding badges that look impressive but do not communicate useful trust.
  • Using screenshots or diagrams without alt text.
  • Letting translated READMEs drift from the main README.
  • Hiding the quickstart below too much explanation.

Example

User:

Make this repository README more attractive for GitHub.

Agent:

  1. Reads the project structure and existing README.
  2. Identifies the target reader and primary use case.
  3. Adds a strong first screen with badges and a local hero image.
  4. Adds a short quickstart and capability table.
  5. Verifies image paths, links, and commands before reporting completion.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most readme changelog skills give in 840 tokens

Counted across 446 of the 460 authors here whose files we hold, read 2026-09-06

  • Follow Keep a Changelog formatin 24 of 446
  • Collect commits since the last git tagin 15 of 446, across 13 files
  • Omit empty sectionsin 14 of 446
  • Put breaking changes first with migration stepsin 14 of 446
  • Include migration guidance for breaking changesin 11 of 446, across 10 files
  • Categorize commits by conventional commit prefixin 11 of 446
  • Mark breaking changes prominentlyin 10 of 446
  • Prepend the new entry to CHANGELOG.mdin 9 of 446
  • Highlight breaking changes with migration notesin 8 of 446, across 7 files
  • Classify changes into Keep a Changelog categoriesin 8 of 446, across 7 files
  • Group related commits into single entriesin 8 of 446
  • Write the changelog from commitsin 8 of 446

Said here and by no other author read

  • identify the primary reader before writing
  • write a first screen with name, value proposition, and badges
  • state what it does, who it serves, why different
  • provide a short copyable quickstart with expected output
  • list key directories, features, or modules
  • add screenshots, diagrams, badges, or examples as proof

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.