readme-maintainer

Produce README files that are fast to skim, technically accurate, and explicit about quality signals. Prefer omission over guessing.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "readme-maintainer" with this command: npx skills add anand-testcompare/scripts-prompts-config/anand-testcompare-scripts-prompts-config-readme-maintainer

README Maintainer

Goal

Produce README files that are fast to skim, technically accurate, and explicit about quality signals. Prefer omission over guessing.

Workflow

  • Collect evidence-backed facts.

  • Verify anything unclear before writing badges or version numbers.

  • Draft or update README with a compact section order.

  • Surface test layers and CI status immediately.

  • Run a final accuracy pass.

  1. Collect Facts

Run the bundled scanner first:

python ~/.codex/skills/readme-maintainer/scripts/readme_facts.py --repo . --format markdown > /tmp/readme-facts.md python ~/.codex/skills/readme-maintainer/scripts/readme_facts.py --repo . --format json > /tmp/readme-facts.json

Use /tmp/readme-facts.json as the source of truth for detected runtime, tools, deployment, external services, API surface, and test/CI gaps.

  1. Verify Badge Inputs

For every versioned badge, confirm:

  • Exact source file is known.

  • Version precision is exact or clearly range-based.

  • Badge value matches the repo state now.

If exact version is unknown, do not invent a number. Use an unversioned badge or omit it.

  1. Use a Compact README Shape

Keep the document short and stable. Default order:

  • Title + one-line summary.

  • Badge block (grouped).

  • What it does (3-7 bullets).

  • How it works (short architecture summary).

  • Quick start.

  • Testing and CI matrix.

  • Deployment and external services.

  • Copyright and license.

If the project has callable APIs, include a short API section and always include an e2e api line in the testing matrix.

  1. Testing and CI Must Be Up Front

Use this table near the top half of README:

Layer Present Tooling Runs in CI

unit yes/no tool list or none

yes/no

integration yes/no tool list or none

yes/no

e2e api yes/no tool list or none

yes/no

e2e web yes/no tool list or none

yes/no

Rules:

  • Show all four rows every time.

  • Keep no visible; do not hide missing layers.

  • If APIs are exposed and e2e api is missing, call it out in plain text immediately below the table.

  1. Badge Rules
  • Group badges by purpose, not alphabetically: stack, tooling, testing/ci, deploy/license.

  • Keep badge count practical; drop low-value badges.

  • Prefer official docs links for each badge target.

  • Include license badge only when a license file or explicit license metadata exists.

Use references/readme-patterns.md for compact badge grouping and section templates.

  1. Final Quality Gate

Before finalizing README changes:

  • Re-run readme_facts.py if code changed during editing.

  • Confirm every numeric badge value is source-backed.

  • Call out missing tests/CI explicitly.

  • Include copyright and license lines when available.

  • Keep prose skimmable: short paragraphs, bullets, and tables.

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

openai-image-generation

No summary provided by upstream source.

Repository SourceNeeds Review
General

convex-delete-deployments

No summary provided by upstream source.

Repository SourceNeeds Review
General

readme-maintainer

No summary provided by upstream source.

Repository SourceNeeds Review
General

ll-feishu-audio

飞书语音交互技能。支持语音消息自动识别、AI 处理、语音回复全流程。需要配置 FEISHU_APP_ID 和 FEISHU_APP_SECRET 环境变量。使用 faster-whisper 进行语音识别,Edge TTS 进行语音合成,自动转换 OPUS 格式并通过飞书发送。适用于飞书平台的语音对话场景。

Archived SourceRecently Updated