Index of every page on this site: llms.txt
All skills
haorantang97 avatar

/write-readme

@d6c47b1
by Lab 305haorantang97/labkit8 stars
Run it yourself

Write or renovate a repository README that explains its purpose, helps readers choose and use its tools, and matches its visual identity. Use for a new repository introduction or an existing landing-page improvement; preserve a narrow edit's scope.

  • 1 file
  • 3.1 KB
  • License
  • Updated last month
  • GitHub
Use this Skill: https://skilld.dev/gh/haorantang97/labkit/write-readme

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈66 tokens for metadata: the name and description. ≈706 when used: this file.

Description uses 3.3% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈263 characters, or 3.3%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Write the repository introduction

Read the repository presentation standard. Inspect actual entrypoints, human documentation, installation paths, licenses and tests before describing capabilities. Infer the repository name, audience and value from available evidence; ask only when an unresolved choice changes the result.

Make the first screen coherent: repository hero, language switcher, truthful badges, concise navigation when useful, and a concrete statement of what the reader can do. A collection's name must remain distinct from each child skill. Use the visual-assets module to replace unfinished artwork before delivery.

Choose structure by purpose. A single tool can lead directly into setup and an example. A collection should help readers choose from public tools before showing installation. A system needs a truthful component/flow map and explicit setup dependencies. Do not force an installation command within a fixed number of lines or make every repository use the same section order.

For a collection, map user tasks to named entrypoints and tangible outputs. Count genuine public tools, not nested SKILL.md files. Use a diagram when relationships benefit from explanation; independent tools should not appear to be a mandatory pipeline. Root documentation introduces and routes; component documentation holds detailed operating instructions.

Give a concrete usage example, meaningful limitations and validation scope. Include the most useful selection/setup FAQs for a substantial page. Links to contribution guidance and actual licenses should resolve. Preserve component-license exceptions and attribution.

Only document installation methods supported by current files and host guidance. Follow the installation boundary: complete-folder installation means the selected skill directory and its dependencies. Point commands at that directory, keeping repository presentation outside the install. Match update instructions to the installed form; a copied folder is not a Git checkout. Do not claim that copying a file activates a hook, sets a model, installs external services or verifies host behavior.

Keep existing language choices. For bilingual repositories, produce complete equivalent editions, synchronize commands and facts, and adapt anchors and diagram labels to each language. A short summary is not a full translation. Fix obsolete names and stale migration/status claims while renovating.

Before delivery, check links and anchors, reconcile the catalogue with the current package tree, inspect rendered assets and the assembled page, and apply the presentation acceptance checks. Keep draft, pushed branch, merged default branch and verified live page as distinct states.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at d6c47b1. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Steadyupdated last month

README badge

README badge for haorantang97/labkit/write-readme