一键导入
create-readme
Create a comprehensive, well-structured README.md for the current project. Use when the user wants to create, write, or refine a project's README.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Create a comprehensive, well-structured README.md for the current project. Use when the user wants to create, write, or refine a project's README.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | create-readme |
| description | Create a comprehensive, well-structured README.md for the current project. Use when the user wants to create, write, or refine a project's README. |
| allowed-tools | Read Write Glob Grep Bash(ls *) Bash(cat *) Bash(git remote *) Bash(git log *) |
You're a senior expert software engineer with extensive experience in open source projects. You always make sure the README files you write are appealing, informative, and easy to read.
Take a deep breath, review the entire project and workspace, then create a comprehensive and well-structured README.md at the project root.
Before writing anything, gather the facts the README must reflect:
Audience & README type — decide who reads this README and what it must primarily enable, then order the document around that intent:
Most projects are a blend; pick the dominant intent and order sections around it.
Identity & purpose — read existing README.md (refine, don't blindly overwrite), package.json / pyproject.toml / Cargo.toml / go.mod, and top-level docs to learn the name, one-line purpose, and description.
How it runs — detect install/build/dev/test commands, entry points, and runtime/version requirements.
Features & usage — infer the main capabilities and the primary usage flow (CLI flags, API, config options, env vars) from source and config.
Assets — look for an existing logo or icon (e.g. logo.*, icon.*, files under assets/, docs/, .github/). If one exists, use it in the header.
Repository — use git remote -v to resolve the repo URL for badges and links.
Never invent commands, features, or badges you cannot verify from the project.
The list below is a menu of sections ordered by a typical priority, distilled from high-quality open source READMEs — not a rigid template. Reorder, drop, and rename sections to match the README type from step 1, and lead with what the primary audience needs first. For a consumption-focused project, that usually means Installation/Getting Started and Usage come right after the overview, and developer/build details are minimized or omitted. Omit any section that does not apply rather than padding it.
Header — project title. If a logo/icon exists, place it centered above the title (<div align="center">…</div>). A short tagline directly under the title.
Badges — a single horizontal row directly under the header, only for things you can verify: build status, package version, runtime/language version requirement, code style, license. Use a consistent style (e.g. ?style=flat-square). Do not fabricate badges.
Include GitHub repository badges (stars and issues), substituting the actual <owner>/<repo> resolved from git remote -v:
[](https://github.com/<owner>/<repo>/stargazers)
[](https://github.com/<owner>/<repo>/issues)
Demo / screenshot — if a demo GIF, screenshot, or animated preview exists in the repo, show it early. Skip if none exists; do not invent a path.
Overview — 1–3 short paragraphs: what the project is, the problem it solves, and who it's for.
Features — a concise bullet list of the key capabilities.
Getting Started / Installation — how a user gets the project working. Interpret this broadly per the README type: for a consumption-focused project this is how to add and install the thing (e.g. adding a marketplace, installing a plugin, pulling an image), including any install options/scopes shown as a table; for a library it's the package install plus prerequisites (runtime version, accounts, keys); for a buildable project it's clone + install + build/dev. Cover only what the audience actually needs.
Usage — the primary flow with copy-pasteable, syntax-highlighted code blocks. Document commands / flags / options / config in a table when there are several. Progress from a basic example to advanced topics.
Examples — concrete, realistic examples that show the project in action (when helpful).
Repository layout / Project structure (optional) — a directory tree with brief annotations, for projects where the layout matters to the reader (collections, monorepos, templates).
Extending / Adding to the project (optional) — steps for how a user adds a new unit (plugin, recipe, module, entry) when doing so is a core use case.
Catalog / Available items (optional) — a table listing what the project offers (e.g. available plugins, commands, presets) when the project is a collection.
Resources / Links — pointers to deeper docs, related projects, or upstream references (only real ones).
Use GFM (GitHub Flavored Markdown) throughout: fenced code blocks with language hints, tables, task lists, and reference links.
Use GitHub admonition syntax where appropriate, with the right type:
> [!NOTE] — neutral clarifications.> [!TIP] — optional, helpful shortcuts.> [!IMPORTANT] — information the user must not miss.> [!WARNING] — consequences of getting something wrong.Example:
> [!NOTE]
> This project requires Node.js >= 20.
Keep a professional yet approachable tone: clear, action-oriented, and concise. Favor short paragraphs and scannable lists.
Heading hierarchy: # for the title only, ## for major sections, ### for subsections.
Use inline code for commands, filenames, flags, and identifiers.
Do not overuse emojis. A single header icon or sparse, purposeful accents are fine; avoid emoji on every bullet.
Keep it concise and to the point — prefer fewer, denser sections over exhaustive ones.
License, Contributing, Code of Conduct, Changelog, or Security sections — those belong in their own dedicated files. You may link to them if the files exist, but do not write their content into the README.plugin.json, package.json, etc.) or a dedicated CHANGELOG. Inlining them creates a maintenance burden and causes the README to go stale.Write the file to the project root and report:
Cross-check an Epic's user stories, PBIs, and acceptance criteria against the actual implementation in its repo, and propose fixes for mismatches, missing/unfinished descriptions, and unclear wording. Use when the user wants a consistency check between backlog and code, 整合性チェック, or asks whether the backlog still reflects what is really implemented.
Investigate a spike task from an Epic's backlog using the Epic's dedicated repo clone, and save a findings report the team can decide from. Use when the user mentions a spike, スパイク調査, or wants a technical question from the backlog researched against the actual codebase.
Load a PBL Epic's working context — backlog snapshot, repo sync data, and the Epic summary — refreshing stale data first, so the conversation can start from the Epic's current state. Use when the user wants to load/open an Epic, asks "where are we on <Epic>", Epicをロード, or starts a session about a specific Epic's status.
Generate an HTML report of an Epic's currently-implemented spec and its upcoming planned features, reconciled from the backlog snapshot and the repo's actual state. Use when the user wants a "what exists now vs what's coming" summary, 仕様まとめ, or a feature-status overview for stakeholders.
Read or update the product backlog on monday.com. Use when the user asks about backlog items, monday.com boards, tickets, or tasks tracked in monday.com.
Generate a sprint-review HTML document for a PBL Epic — done items with demo pointers, sprint metrics, and an agenda — from the monday.com board and its Drive snapshot. Use when the user wants sprint review material, a demo agenda, or to prepare for a sprint review / スプリントレビュー.