New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Create Doc

by simonliu-ai-product in simonliu-ai-product/open-doc

A skill your agent uses when the user wants to create, draft, author, or generate a new document, report, whitepaper, proposal, memo, or spec in this open-doc repo.

MITAuto-check passedDocuments & Office

Install Create Doc

skills CLI
$ npx skills add simonliu-ai-product/open-doc --skill create-doc -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install simonliu-ai-product/open-doc create-doc --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/simonliu-ai-product/open-doc.git skills-src && mkdir -p .claude/skills && cp -r skills-src/packages/core/skills/create-doc .claude/skills/create-doc && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
create-doc
GitHub stars
108
Token cost
~2.4k tokens
SKILL.md length
1,296 words
Files
1
Skills in repo
7
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when the user wants to create, draft, author, or generate a new document, report, whitepaper, proposal, memo, or spec in this open-doc repo.

  • Works in 9 steps: Pick a theme → Gather the substance first → Clarify the shape (MUST ask before… → …
  • The user wants to create
  • SKILL.md covers Step 0 — Pick a theme, Step 1 — Gather the substance…, Step 2 — Clarify the shape… and Step 3 — Pick a doc id, plus 5 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Create Doc is an agent skill from simonliu-ai-product/open-doc. Use this skill when the user wants to create, draft, author, or generate a new document, report, whitepaper, proposal, memo, or spec in this open-doc repo. Triggers on phrases like "write a report about X", "draft a proposal", "make a whitepaper", "create a document", "write up the Q3 results", or when the user asks to add content under docs/. Do NOT use for editing the framework itself — only for authoring content inside docs/<id/.

Its SKILL.md is about 2.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Documents & Office, covering Report writing. The repository describes itself as: The document framework built for agents — write reports as React, get real A4 pages, an outline, page numbers, and a PDF. The licence is MIT.

When your agent uses it

  • The user wants to create
  • Generate a new document
  • Spec in this open-doc repo
  • Phrases like write a report about X

Example prompts

  • “write a report about X”
  • “draft a proposal”
  • “make a whitepaper”
  • “/create-doc”

Workflow steps

9 steps, taken from the step headings in SKILL.md.

  1. Pick a theme
  2. Gather the substance first
  3. Clarify the shape (MUST ask before writing code)
  4. Pick a doc id
  5. Outline before code
  6. Commit to a visual direction
  7. Write docs//index.tsx
  8. Self-review
  9. Hand off

What it can do on your machine

Read from SKILL.md and the folder at commit af6095d. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Create Doc loads about 2.4k tokens when it runs. Until then it costs about 113 tokens; SKILL.md has 1,296 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~113
When it runs · the whole SKILL.md, loaded when a task matches
~2.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from simonliu-ai-product/open-doc at commit af6095d, republished under its MIT licence (© simonliu-ai-product). 1,296 words, ~2,375 tokens.

Download SKILL.mdSave it as .claude/skills/create-doc/SKILL.md (or your agent's skills folder).
name
create-doc
description
Use this skill when the user wants to create, draft, author, or generate a new document, report, whitepaper, proposal, memo, or spec in this open-doc repo. Triggers on phrases like "write a report about X", "draft a proposal", "make a whitepaper", "create a document", "write up the Q3 results", or when the user asks to add content under `docs/`. Do NOT use for editing the framework itself — only for authoring content inside `docs/<id>/`.

Create a document in open-doc

This skill owns the workflow for drafting a new document. The technical reference — file contract, page canvas, print type scale, vertical budget, tables, TOC — lives in the doc-authoring skill. Read it before writing code; don't duplicate its rules here.

You only write files under docs/<id>/. Never modify package.json, open-doc.config.ts, or existing documents.

If the user already has the content written as Markdown, don't retype it into JSX. open-doc import <file.md> --id <id> produces a real document — flow() body, cover, contents, local images copied into the document's assets — which you then refine. Steps 0–2 still apply for the parts the import cannot know (theme, page size, visual direction).

If a template fits the kind of document, start from it. open-doc templates lists them — built-in (report, proposal, meeting-notes, letter, certificate, blank) and the workspace's own under templates/<name>/ — with the name to pass on. open-doc new <id> --template <name> --title "<title>" copies one into docs/<id>/, retitled and dated today, and you then replace its stand-in content with the real substance (Steps 1–2 still apply). Over MCP, the same is list_templates and create_from_template.

Step 0 — Pick a theme

List files under themes/. If any theme markdown exists (anything other than README.md), call AskUserQuestion with each theme id as an option plus a final "no theme — design from scratch" option. (AskUserQuestion holds at most 4 options — with 4+ themes, offer the 3 most relevant plus "no theme"; the auto-added "Other" lets the user name an omitted one.)

  • If the user picks a theme: read themes/<id>.md end-to-end. Its palette, typography, page setup, and paste-ready components are now authoritative — copy them into the document, and set meta.theme: '<id>' in index.tsx. Skip the visual direction question in Step 2 (the theme already commits to one) and restate the theme name so the user can correct course. Page size comes from the theme's frontmatter unless the user overrides it.
  • If the user picks "no theme", or themes/ has no theme files: continue unchanged.

Step 1 — Gather the substance first

A document is judged on content, not layout. Before anything else, establish:

  • Topic and purpose — what decision or action should this document produce?
  • Audience — executives (lead with the recommendation), engineers (lead with the mechanism), clients (lead with the outcome and the price).
  • Source material — does the user have data, a draft, notes, a repo, an existing doc? Ask for it. Never fabricate figures, quotes, citations, or customer names.

If the request is thin ("write me a report"), make a separate AskUserQuestion call for topic, audience, and source material before the style questions below. If the topic is already clear, restate your reading of it in the next call so the user can correct course.

Step 2 — Clarify the shape (MUST ask before writing code)

Ask these in a single AskUserQuestion call (multi-question form). Skip a question only when the user's message already answers it unambiguously — and restate the assumption if you skip.

  1. Document type — propose 3 concrete types that fit this topic, each with what it implies structurally. Not bare labels:
    • "our Q3 infrastructure numbers" → internal review (summary, metrics, incidents, actions) · exec brief (2 pages, recommendation-first) · post-incident deep dive (timeline, root cause, remediation)
    • "our new product for enterprise buyers" → whitepaper (problem, approach, evidence, references) · sales proposal (scope, deliverables, timeline, price) · one-pager (positioning + proof + CTA) Mark the best fit "(Recommended)".
  2. Length — offer brackets: 2–4 pages (brief), 5–10 (standard report), 11–25 (deep dive). The auto-added "Other" covers custom counts.
  3. Page size — A4 portrait (default) · A4 landscape (data-heavy, wide tables) · B4 or A3 (plans, posters, very wide tables). The only sizes are A4, B4 and A3, each portrait or landscape. Skip if the user already said.
  4. Visual direction — 3 options tailored to the audience, each naming a palette + typographic cue: e.g. corporate neutral (near-white, single blue accent, sans throughout) · editorial serif (serif headings, generous leading, hairline rules) · technical mono-accent (mono labels, dense tables, monochrome + one signal color).

Ask about brand colors, a logo, or required sections only if still unclear afterwards.

Step 3 — Pick a doc id

kebab-case, short, descriptive: q3-infra-review, acme-proposal-2026, auth-migration-rfc. Check docs/ for collisions.

Show full SKILL.md (613 more words)Show less

Step 4 — Outline before code

Write the page plan first — one line per page, with its role. Typical structures:

TypePage plan
ReportCover · Contents · Executive summary · Method · Findings (n pages) · Risks · Recommendations · Appendix
WhitepaperCover · Abstract · Problem · Approach · Evidence · Comparison · Conclusion · References
ProposalCover · Summary · Scope · Deliverables · Timeline · Pricing · Terms · Next steps
RFC / specSummary · Motivation · Design · Alternatives · Migration · Open questions
MemoHeader block · Context · Recommendation · Rationale · Next steps

Rules for the plan:

  • Page 1 is the cover; if the doc is over ~6 pages, page 2 is <TableOfContents />. Both are fixed pages; everything after them belongs in one flow() section so the framework paginates the body.
  • The executive summary must stand alone — a reader who reads only that page should know the conclusion and the ask.
  • One section per page unless a section is genuinely two paragraphs long.
  • Plan the body as a sequence of sections, not as a page count — flow() decides how many pages that becomes. Only estimate the vertical budget for the fixed pages (cover, contents, dividers); see doc-authoring → references/pagination.md.

Show the user the page plan and the estimated page count before writing the file if the document is over ~8 pages — restructuring a 20-page doc after the fact is expensive.

Step 5 — Commit to a visual direction

Declare a top-level export const design: DesignSystem and consume var(--od-*) everywhere. Print constraints (white background, ≥13px body, one accent) are in references/design-system.md — read it before choosing the palette. If a theme was picked in Step 0, copy its design const verbatim instead of inventing one.

Keep the const a plain object literal: the dev UI's Design panel parses and rewrites it, and a spread or a value read from another constant makes the document untweakable (see "Writing for the Design panel" in doc-authoring).

Define the shared page shell, heading styles, and the running footer once as local constants/components at the top of index.tsx, then reuse them on every page. Copy-pasted page styling is how documents drift.

Step 6 — Write docs/<id>/index.tsx

Read doc-authoring first — file contract, canvas, type scale, headings/outline, TOC, page numbers, and a starter template are all there.

While writing:

  • Real <h1>/<h2>/<h3> for every section title, so the outline and TOC populate.
  • <TableOfContents /> for the contents page. Never hand-write one.
  • useDocPageNumber() / useDocPageCount() for the footer. Never hardcode.
  • Put body content in one flow(<>…</>, { footer: Footer }) section; mark captions data-od-keep-with-previous. Only run budget math for fixed pages.
  • Where the user must supply data, leave <ImagePlaceholder hint="…"> or an explicit TODO: in the copy — never invent numbers.

Step 7 — Self-review

Run open-doc check <id> first. You cannot see the sheets you produced; it renders them at true page size and reports clipped content, blank pages, stranded headings, and unreadable type, each with a source location. Fix every error before moving on. (Driving the MCP server instead? check_layout, and render_page when you need to look at a sheet.)

Then run the checklist in doc-authoring ("Self-review before finishing"), and re-read the prose once as a reader: does the summary state the conclusion? Does every claim have a source?

Step 8 — Hand off

Tell the user:

  • The doc id and file path.
  • That the dev server hot-reloads — open http://localhost:5273/d/<id> (or refresh the home page).
  • Every placeholder and TODO: you left, and what data each one needs.
  • That "Export PDF" in the toolbar prints at the true page size, and "HTML" downloads a self-contained copy — or open-doc export <id> for the same files without a browser.
  • That the Design button live-tweaks the palette and type scale and writes the result back to the source, and that images they want to drop in go through the Assets page.

Don't run the dev server yourself unless asked.

© simonliu-ai-product, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in packages/core/skills/create-doc of simonliu-ai-product/open-doc.

Open the folder on GitHubat commit af6095d

Compare with similar skills

Create Doc next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Create Doc compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Create Doc this skillsimonliu-ai-product/open-doc108—~2.4kAutomated safety check: PassMIT
Graphic EbookVarnan-Tech/opendirectory674—~5kAutomated safety check: PassMIT
Make Photo Stamp ArchiveDlcccc71913/skill-make-photo-stamp-archive367—~1.3kAutomated safety check: PassMIT
Make Tape Collagesherlyryn/make-tape-collage118—~4.7kAutomated safety check: PassMIT
Ky Markdown RebuilderKyrieCheungYep/ky-markdown-rebuilder117—~5.7kAutomated safety check: PassNone
HTML Summarytestdouble/han279—~2.9kAutomated safety check: PassMIT

Similar skills

  • Graphic Ebook

    Varnan-Tech/opendirectory

    Creates professionally designed B2B SaaS e-books in HTML + CSS, exported as print-ready PDF.

    674 GitHub stars~5k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed

  • Make Photo Stamp Archive

    Dlcccc71913/skill-make-photo-stamp-archive

    Transform one or more supplied photos into clean direct-splice archival artworks that pair a faithfully preserved photograph with a warm-white paper panel containing a compact hand-pressed custom…

    367 GitHub stars~1.3k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed

  • Make Tape Collage

    sherlyryn/make-tape-collage

    Transform a supplied photo or text description into a clean, tactile tape-collage raster artwork, or pair a faithfully preserved borderless photo print with a spacious warm-white paper panel…

    118 GitHub stars~4.7k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed

  • Ky Markdown Rebuilder

    KyrieCheungYep/ky-markdown-rebuilder

    Rebuild visual documents into reliable Markdown by combining text extraction with page or screenshot alignment.

    117 GitHub stars~5.7k tokensUpdated 3 mo ago
    Documents & OfficeAuto-check passed

  • HTML Summary

    testdouble/han

    Convert a stakeholder summary markdown file into a single self-contained HTML executive report — bottom line and decision asks up front, supporting detail later — styled with a Test Double-derived…

    279 GitHub stars~2.9k tokensUpdated 8 days ago
    Documents & OfficeAuto-check passed

  • Summarize

    reysu/ai-life-skills

    Summarize any content (YouTube video, article, whitepaper/PDF, podcast episode, book chapter, etc.) into a rich Obsidian note with section-by-section breakdowns, wikilinks to all technical concepts…

    269 GitHub stars~7.7k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed

More from simonliu-ai-product/open-doc

  • Apply Comments

    simonliu-ai-product/open-doc

    A skill your agent uses when the user asks to apply, process, or clear the comments they left in the open-doc inspector — phrases like "apply the comments", "apply my edits", "I left notes on the…

    108 GitHub stars~948 tokensUpdated 2 days ago
    Auto-check passed

  • Doc Authoring

    simonliu-ai-product/open-doc

    Technical reference for writing or editing open-doc pages — file contract, the A4/B4/A3 page canvas, print type scale, the vertical budget that decides where a page breaks, tables, charts, table of…

    108 GitHub stars~6.8k tokensUpdated 2 days ago
    Auto-check passed

  • Current Doc

    simonliu-ai-product/open-doc

    Resolve which document, page, and (optionally) selected element the user is currently viewing in the open-doc dev server.

    108 GitHub stars~1.8k tokensUpdated 2 days ago
    Auto-check passed

  • Doc Runtime Patterns

    simonliu-ai-product/open-doc

    Implementation patterns for the @open-doc/core runtime — the split between the viewer and the published bundle, virtual modules, the flow pipeline, the ops layer, dev-only plugins, and the…

    108 GitHub stars~2.1k tokensUpdated 2 days ago
    Auto-check passed

  • Print Layout Review

    simonliu-ai-product/open-doc

    Reviews page layout, typography, and pagination code in open-doc against a print craft bar — the sheet is the deliverable, not the screen.

    108 GitHub stars~2k tokensUpdated 2 days ago
    Auto-check passed

  • Viewer UI Guidelines

    simonliu-ai-product/open-doc

    Design and accessibility rules for the open-doc viewer chrome — the browser shell, sidebars, thumbnail rail, outline, assets and design panels, inspector overlay, and menus.

    108 GitHub stars~1.6k tokensUpdated 2 days ago
    Auto-check passed

Categories

Questions about Create Doc

What does Create Doc do?

A skill your agent uses when the user wants to create, draft, author, or generate a new document, report, whitepaper, proposal, memo, or spec in this open-doc repo. Create Doc is an agent skill from simonliu-ai-product/open-doc. Use this skill when the user wants to create, draft, author, or generate a new document, report, whitepaper, proposal, memo, or spec in this open-doc repo.

When should I use Create Doc?

Create Doc fits situations like: the user wants to create; generate a new document; spec in this open-doc repo; phrases like write a report about X.

How do I install Create Doc in Claude Code?

Run `npx skills add simonliu-ai-product/open-doc --skill create-doc -a claude-code`. Or copy the skill folder (packages/core/skills/create-doc in simonliu-ai-product/open-doc) into .claude/skills/create-doc in your project. Claude Code loads it when a task matches its description.

How do I install Create Doc in Codex?

Run `npx skills add simonliu-ai-product/open-doc --skill create-doc -a codex`. Or copy the skill folder (packages/core/skills/create-doc in simonliu-ai-product/open-doc) into .agents/skills/create-doc in your project. Codex loads it when a task matches its description.

Can I use Create Doc in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add simonliu-ai-product/open-doc --skill create-doc -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/create-doc, .gemini/skills/create-doc, .github/skills/create-doc and .opencode/skills/create-doc in your project.

What does Create Doc need to run?

SKILL.md names no scripts, command-line tools or credentials: Create Doc is instructions for the agent only.

Does Create Doc access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Create Doc safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Create Doc use?

Create Doc is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Create Doc use?

About 2.4k tokens (SKILL.md is roughly 9.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Create Doc?

Skills that share tags, products or a category with Create Doc: Graphic Ebook (Varnan-Tech/opendirectory, 674 stars), Make Photo Stamp Archive (Dlcccc71913/skill-make-photo-stamp-archive, 367 stars), Make Tape Collage (sherlyryn/make-tape-collage, 118 stars) and Ky Markdown Rebuilder (KyrieCheungYep/ky-markdown-rebuilder, 117 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Create Doc?

simonliu-ai-product (a GitHub organization) maintains it in simonliu-ai-product/open-doc, which has 108 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on October 7, 2026.

Source: simonliu-ai-product/open-doc on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.