Build With Univer

Organization

dream-num

build-with-univer

Build applications with Univer Office SDK, covering the Web SDK (browser Office editors, headless Sheets and Docs in Node), Server SDK (self-hosted OT collaboration, storage, auth, Office import/export), AI SDK (agent-operated Office CLIs, reviewable Worktrees), and Binding SDK (plain HTML pages bound to live Sheet data). Use when starting or designing any Univer-based app or choosing Univer packages, including when the user describes a spreadsheet, document, collaboration, or AI-office need without naming a Univer package.

Overview

Publisher	dream-num
Repository	`skills`
Skill name	`build-with-univer`
Stars	79
Forks	7
Bundled files	7
License	Apache-2.0
Links	Skill source Repository Homepage

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
7 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by dream-num on GitHub. Read the source before you install it.

Installation

Install the Build With Univer AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/dream-num/skills/tree/main/skills/build-with-univer
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/dream-num/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/skills/build-with-univer .claude/skills/build-with-univer

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Build With Univer in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Build With Univer on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Build With Univer is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Build with Univer

Univer Office SDK is a family of npm packages, not a single library. Pick the SDK by what you are building, then follow its example repository where one exists. Do not guess APIs from memory: Univer changes quickly and older material is widely mis-remembered.

Pick the SDK

You are building	SDK	Packages	Start from
An embedded spreadsheet, document, or slide editor in a web app	Web SDK	`@univerjs/`, `@univerjs-pro/`	references/web-sdk.md
Generate or modify Sheet or Doc data, or calculate formulas in Node.js	Headless Web SDK	`@univerjs/*` without editor UI	references/web-sdk.md
Real-time multi-user editing with your own auth and storage	Server SDK + Web SDK client	`@univerjs-pro/collaboration-*`	references/server-sdk.md, `quick-start` example
Server-side Office import/export (xlsx, docx, pptx)	Server SDK	`@univerjs-pro/exchange-node`	references/server-sdk.md, `exchange` example
Your own CLI or tool that lets agents read and edit Office content	AI SDK	`@univer-cli/*`	references/ai-sdk.md, `01-content-operations` example
A plain HTML/CSS/JS page showing or editing live Sheet cells	Binding SDK	`@univerjs-labs/*`	references/binding-sdk.md (installed packages' docs)

Real products often combine them. A collaborative editor is Web SDK in the browser plus Server SDK behind it. An HTML dashboard over shared data is Binding SDK on top of a Server SDK backend. Running in Node does not by itself select Server SDK or AI SDK: use Headless Web SDK for document processing, Server SDK for collaboration/storage or Office import/export, and AI SDK for agent tools.

The four SDKs

Web SDK is the engine. It runs in the browser or headless in Node and covers six products: Sheets and Docs (open source), Slides, Boards, Bases and PDFs (Pro). Choose the runtime before assembling plugins: browser editors need UI; headless Node processes document data without an editor UI. Follow each product's runtime guide rather than assuming Node support. For new Web SDK integrations, default to plugin mode; preserve existing presets or an explicit request for them. All programmatic control goes through the Facade API.
Server SDK is a set of Node.js packages you assemble inside your own backend: transport, endpoint, service and a database adapter, with an OT collaboration engine and history, comment and Worktree services. Your application owns users, permissions, storage and deployment.
AI SDK builds Office CLIs for agents. Your app owns the Commander program and its commands; the SDK supplies headless runtimes, content inspection and execution, offline Facade API search, rendering and layout lint, and Worktree-based review.
Binding SDK connects ordinary HTML pages to Sheet data with data-univer-* attributes. Pages render in a sandboxed iframe connected to a host, and need a Server SDK collaboration backend.

Working principles

Read the target project first: framework, package manager, and the installed @univerjs/*, @univerjs-pro/* and @univer-cli/* versions. Preserve what is installed unless asked to upgrade.
For anything new, read in this order and then implement: the relevant docs page, the matching example repository in full (where the SDK has one), then the README of each package you import. Never invent APIs; if an export is not in the installed declarations or the docs, it does not exist. Clone example repositories into a git-ignored directory inside the project (for example .ref/) and remove it before you finish. If a repository is not reachable, the installed packages' README.md and docs/ are the reference.
Pick versions at build time, not from memory: check the project's lockfile, then npm dist-tags. Keep every package in one release group on the same exact version. See references/versions-and-license.md, which also covers the @univerjs-labs registry and license wiring.
Copy an example's wiring rather than recombining packages from scratch. Then replace its demo shortcuts (fixed user, always-allow authorization, in-memory storage) with real ones.
Verify in proportion to the change: build, run, and for collaboration open two independent browser contexts. Run headless examples in plain Node. Before declaring a feature unsupported or adding a workaround, check the runtime's installed packages, plugin registrations and Facade imports, then verify the native API. A Facade method can exist without its feature plugin.
A missing license never blocks a proof of concept. Everything still runs; Pro surfaces show watermarks or restrict some operations. Build first, add the license when one is available.
Check references/avoid.md before recommending anything from older docs, blog posts or your own memory: superseded servers, packages and API names are listed there.

Docs for agents

https://docs.univer.ai/llms.txt is the index; llms-full.txt is the full bundle.
Every docs page has a Markdown twin: append .md to the URL.
Section indexes: /guides/llms.txt (Web SDK), /server/llms.txt, /ai/llms.txt, /reference/llms.txt.
https://docs.univer.ai/server/architecture.md is the canonical page for choosing entry points.

Topics beyond the first build

Look these up in the docs instead of guessing. Every URL has a Markdown twin (append .md).

Topic	Where
Facade API for a product	`https://docs.univer.ai/guides/<product>/getting-started/facade`
Pro products, advanced formula engine, permissions	`/guides/sheets/features/advanced-formula`, `/guides/sheets/features/core/permission`
Custom UI components, plugins, commands	`/guides/sheets/ui/components` and the package reference at `/reference/llms.txt`
Themes and dark mode	`/guides/fundamentals/ui/themes`
Web Workers, i18n, headless Node	`/guides/sheets/getting-started/{worker,i18n,node}`
Server-side collaboration and Office import/export	`/server/llms.txt`
Agent-operated Office content	`/ai/llms.txt`

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Build With Univer AI skill do?

Why use Build With Univer on TypingMind?

Because you install it once and use it with any model. Build With Univer is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Build With Univer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/dream-num/skills/tree/main/skills/build-with-univer. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Build With Univer?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Build With Univer?

As many as you like. As long as a model supports skills, you can use Build With Univer with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Build With Univer AI skill free?

Yes. It is published on GitHub by dream-num under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Comps Analysis

NousResearch

OrganizationPopular

Build comparable-company valuation workbooks in Excel.

Lbo Model

Build leveraged buyout workbooks with IRR/MOIC in Excel.

Pptx Author

Build PowerPoint decks headless with python-pptx.

Pokemon Player

Play Pokemon via headless emulator + RAM reads.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

Build with Univer

Pick the SDK

The four SDKs

Working principles

Docs for agents

Topics beyond the first build

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now