hashgraph-online/awesome-codex-plugins / plugins/alpoxdev/hypercore/skills/hyper-documentation
hyper-documentation
Flagged[Hyper] Use this skill when the user asks to create, refactor, or update a documentation artifact: a project or package README.md, a repository AGENTS.md with its AGENTS.ko.md mirror, an AI-readable instruction base, runbook, specification, or rule pack, a reusable prompt, prompt template, prompt pack, or prompt eval fixture, or an agent's persistent memory files (MEMORY.md index, memory bank, lessons file) with their write, pruning, and deletion rules. It picks one mode from the requested output and grounds every claim in the project and its sources. Do not use for reusable skill folders, CHANGELOG, CONTRIBUTING, LICENSE, release notes, answer-only research, or implementation work; it never writes CLAUDE.md.
rules/readme/project-discovery.md
View on GitHub# Project Discovery
**Purpose**: Read the project carefully and detect what shape of README it needs before drafting.
## 1. Shape detection signals
Detect the project type before choosing sections:
| Project type | Strong signals |
|---|---|
| **CLI** | `bin` field in `package.json`, `src/cli/*`, single executable script, `Cargo.toml` `[[bin]]`, Python `entry_points`/`console_scripts` |
| **Library** | `main`/`module`/`exports` in `package.json`, `lib/`, `src/index.*`, `Cargo.toml` `[lib]`, Python package without CLI entry |
| **Web app** | `next.config.*`, `vite.config.*`, `nuxt.config.*`, `app/` or `pages/`, framework deps in manifest |
| **Monorepo** | `packages/`, `apps/`, `pnpm-workspace.yaml`, `turbo.json`, `nx.json`, `lerna.json`, root `workspaces` field |
| **Plugin** | `.claude-plugin/`, `agents/`, `skills/`, plugin/marketplace manifest for an editor or platform |
| **Framework** | exposes templates, generators, or a `create-*` command |
| **Docs site** | `docusaurus.config.*`, `astro.config.*`, `mkdocs.yml`, `vitepress`, large `docs/` tree |
| **Service / API** | `Dockerfile`, `serve` script, server entry, route folder, no published library exports |
When signals overlap, prefer the dominant deliverable: a published package without an app entry is a library; a server with a published SDK may be both — document the primary use first and link to the secondary surface.
## 2. Files to inspect first
Read in this order before drafting:
1. Existing `README.md` (root and package-level if monorepo).
2. `AGENTS.md`, `CLAUDE.md`, `GEMINI.md` for language and convention hints.
3. `LICENSE` or `LICENSE.*`.
4. Manifest: `package.json`, `pyproject.toml`, `Cargo.toml`, `go.mod`, `deno.json`, `pubspec.yaml`.
5. Lockfile to confirm package manager: `pnpm-lock.yaml`, `package-lock.json`, `yarn.lock`, `bun.lockb`, `Cargo.lock`, `poetry.lock`, `uv.lock`.
6. Workspace manifests: `pnpm-workspace.yaml`, `turbo.json`, `nx.json`, `lerna.json`.
7. Entry points: `bin`, `main`, `module`, `exports`, `src/index.*`, `src/cli/*`, framework config files.
8. Scripts: `scripts` field in `package.json`, `Makefile`, `justfile`, `Taskfile.yml`.
9. Plugin manifest: `.claude-plugin/*.json`, `agents/`, `skills/`, marketplace metadata.
10. Top-level folders to map structure: `src/`, `packages/`, `apps/`, `cli/`, `instructions/`, `scripts/`, `docs/`.
Stop reading once shape and primary commands are clear. Do not exhaustively read source files.
## 3. Install command derivation
Choose the install snippet from the actual manifest, never from convention:
- Node: pick the package manager from the lockfile (`pnpm install`, `npm install`, `yarn install`, `bun install`).
- Published library: show consumer install from the manifest `name` field (`npm install <name>`, `pnpm add <name>`).
- CLI bin: show the run path (`npx <name>`, `pnpm dlx <name>`, `cargo install <name>`, `pipx install <name>`).
- Plugin/marketplace: derive from the project's own existing README or plugin manifest. Do not invent marketplace URLs.
- Local-only project (private or unnamed): write the from-source install (`git clone`, then run dev/build).
If the manifest has no `name` or it is private, do not invent one. Use the local install path instead.
## 4. Usage and example derivation
Look for the smallest authentic example:
- **Library**: read the public exports and pick a 3-10 line snippet that uses the most documented function.
- **CLI**: list the top-level commands from the actual command-handler file or captured `--help` output; quote one realistic command.
- **Web app / service**: show the dev command, the URL it serves, and the production build command.
- **Plugin**: show how the plugin is installed and invoked in the host environment, derived from existing project docs.
Mark fabricated or hypothetical examples with `<!-- TODO -->` rather than ship them as real.
## 5. Project map only when needed
Add a "Project structure" section only if:
- the repo has 4+ top-level folders that a new contributor needs to navigate, or
- it is a monorepo with multiple packages worth listing.
Otherwise omit it. The README should orient, not mirror the file tree.
## 6. Language and convention hints
Use these signals to decide whether the README can use the Korean default or must preserve another target language:
- Default to Korean for new README prose.
- If an existing target `README.md` is predominantly non-Korean and the task is a surgical refactor/update, preserve that language for consistency unless the user asked to translate.
- If the user explicitly requests a language, use it. If preservation signals conflict and would materially change the artifact language, ask once; do not silently choose English.
## 7. Discovery output
Before drafting, record a short profile in working notes:
```text
Type: [cli | library | web-app | monorepo | plugin | framework | docs-site | service]
Package manager: [npm | pnpm | yarn | bun | cargo | uv | poetry | go | deno | other]
Entry: [path or command]
Primary scripts: [dev/build/test/lint/release]
README language: [ko default | user-requested | preserve-existing]
Existing README state: [missing | present-accurate | present-stale]
License: [SPDX or "missing"]
```
This profile drives section selection in `rules/readme/section-design.md` and the validation summary in `rules/readme/validation.md`.
## Sources
> No external sources; content checked 2026-09-21.
This rule states the repository-local discovery procedure for README authoring. It makes no external claim, so no external source is cited.
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: 469k | 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: 197k | 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: 197k | 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: 197k | Updated | Claude CodeCodex | Verified | Install |