sidiangongyuan/codex-skills-library / skills/github-project-release
github-project-release
VerifiedUse when preparing or publishing a local project as a clean GitHub repository. Covers new repositories, private staging releases, public README cleanup, updates to existing remotes, and research-code packaging without datasets, checkpoints, papers, or generated artifacts.
| File | Kind | Size |
|---|---|---|
| LICENSE | other | 1.0 KB |
| SKILL.md | skill md | 4.3 KB |
| agents/openai.yaml | config | 302 B |
| references/readme_playbook.md | markdown | 4.1 KB |
| scripts/github_project_release.py | script | 20.9 KB |
references/readme_playbook.md
View on GitHub# README Playbook
Use this guide when creating or polishing a README for a GitHub release repository.
## Goal
Make the repository understandable to a new external reader. The README is not a lab note, handoff
message, rebuttal record, submission checklist, or conversation with the project owner.
## Default Style
- Write in English unless the user explicitly asks for another language.
- Prefer a polished GitHub project-page rhythm: centered title, light badges/links, optional curated
visual, short overview, what is included, setup, run commands, citation, acknowledgements, license.
- Adapt sections to the project. Do not force a fixed template if a section would be empty.
- Keep the first screen concrete: project name, status, authors or organization if relevant, and one
clear sentence about what the project does.
## Required Coverage
Every release README must answer:
- What is this project?
- What problem does it solve?
- What is included in this repository?
- How does a new user install and run the minimal path?
- What datasets/checkpoints/external assets are required but not included?
- How should users cite the project?
- Which upstream repos/papers should also be credited?
If any answer is unavailable, write an honest placeholder such as `Paper link: coming soon.` or
`Pretrained checkpoints will be released separately.` Do not invent facts.
## Top Section
Use a centered block only when it improves the page:
```md
<div align="center">
# Project Name
<p><b>Venue or status, if public</b></p>
<p>
<a href="#quick-start"><img src="https://img.shields.io/badge/Code-Runnable-green" alt="Code"></a>
<a href="CITATION.cff"><img src="https://img.shields.io/badge/Citation-CFF-orange" alt="Citation"></a>
<a href="LICENSE"><img src="https://img.shields.io/badge/License-Academic-lightgrey" alt="License"></a>
</p>
</div>
```
Use only real links. If the paper URL is not public, write `Paper link: coming soon.` in prose
instead of making a fake badge.
## Explaining The Project
- Explain key terms only when needed. Do not over-explain common field terms.
- Avoid marketing claims that are not supported by the paper or code.
- State the minimal runnable scope clearly. If only a subset is released, say so plainly.
- If the project is derived from upstream code, explain the current contribution and the upstream
lineage without blurring them.
## Assets And Results
Allowed when curated and relevant:
- teaser image
- method diagram
- main result table from the paper
- benchmark summary from the paper
- small logo or architecture figure
Do not include:
- raw visualization dumps
- temporary experiment screenshots
- full ablation folders
- logs or metrics copied from training outputs
- large videos or archives
- paper PDFs or camera-ready files
Use paper-confirmed results only. Do not extract numbers from logs unless the user explicitly says
those logs are the release source of truth.
## Paper And Citation
- If the paper/arXiv link is public, link it.
- If not public, write `Paper link: coming soon.`
- Do not upload or link local PDFs, camera-ready files, OpenReview-private material, rebuttals,
review text, or submission instructions.
- Include BibTeX when available. If exact BibTeX is unknown, use a clearly marked placeholder rather
than fabricating venue, title, authors, or year.
## Upstream Credit
For derivative research code, include both:
- `Acknowledgements`: plain-language credit with upstream repo links.
- `Citation`: upstream BibTeX entries when the user or repo provides verified citations.
Do not preserve an upstream README as the front page for the new project. Use it only as source
material for setup commands, dependency notes, license terms, and upstream attribution.
## Red Flags To Remove
Remove or rewrite content that mentions:
- local paths such as `/mnt/...`, `/home/...`, or machine-specific usernames
- instructions written to the project owner
- TODOs from the release-preparation conversation
- camera-ready, rebuttal, review, meta-review, or submission-process details
- private datasets, private checkpoints, or unpublished artifact paths
- claims that code, checkpoints, or papers are included when they are not
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |