HTML to DOCX Setup

This skill adds @turbodocx/html-to-docx to a JavaScript/TypeScript project — a zero-dependency library that converts HTML strings to Word documents without Puppeteer, Chrome, or LibreOffice. It runs in Node.js and in the browser (via the bundled IIFE/ESM builds).

Prefer server-side when possible. Running this server-side is faster, avoids the ~2.4 MB browser bundle, sidesteps polyfill requirements, and keeps sharp (for SVG → PNG conversion) available. Default to a server-side integration whenever the project has any backend (Express/Fastify/Next.js API route/etc.) and only fall back to the browser bundle when the project is genuinely static or the user explicitly asks for client-side generation.

Phase 1: Detect the Project Type

Use Glob to check what kind of project this is:

1. Node.js / npm-based project — package.json exists in the root. This is the preferred path.

- Also check for tsconfig.json to determine if TypeScript is in use. - Detect the package manager from lockfiles: - pnpm-lock.yaml → pnpm - yarn.lock → yarn - bun.lockb → bun - package-lock.json or none → npm - Check package.json for "type": "module" to determine ESM vs CommonJS imports. - If the project has both a server runtime and a client (e.g., Next.js, Remix, Nuxt, SvelteKit), default to the server route/handler/action path — do not put document generation in a client component unless the user explicitly asks. - Proceed to Phase 2 (npm install path).

1. Browser-only project — no package.json, but HTML files exist (*.html) or the user has explicitly said they want to use this in a static page / CDN setup. Before committing to this path, briefly confirm with the user that they don't have a backend they'd rather run this in — server-side is preferred. If they confirm browser is required:

- Skip npm install entirely. The library ships a self-contained browser bundle (dist/html-to-docx.browser.js, ~2.4 MB IIFE) with all dependencies inlined. - Read the Browser Usage section in references/usage.md for the polyfill snippet, the HTMLToDOCX(...) global, and the limitations (no sharp, CORS-restricted remote images, no filesystem). - Drop in a <script src="..."> referencing either a hosted copy or a CDN build, and generate a minimal working example tailored to the user's page. - Skip Phases 2-4 below; the browser path is install-less.

1. Neither — no package.json and no HTML files. Ask the user which environment they're targeting (Node.js, bundler, or static HTML) before proceeding, and recommend server-side as the default.

Phase 2: Install the Package

Run the install command for the detected package manager:

• npm: npm install @turbodocx/html-to-docx
• pnpm: pnpm add @turbodocx/html-to-docx
• yarn: yarn add @turbodocx/html-to-docx
• bun: bun add @turbodocx/html-to-docx

No environment variables are needed — this is a local library with no API keys.

Phase 3: Read Reference

Read references/usage.md from this skill's directory. It contains the full API surface, configuration options, and code examples.

Phase 4: Analyze Codebase and Generate Code

Step 4.1: Explore the project

Use Glob and Read to understand:

• Where source files live (src/, app/, lib/, root)
• What framework is in use (Express, Fastify, NestJS, Next.js, Hono, Koa, or none)
• Existing patterns: how routes are defined, how files are organized, naming conventions
• Whether the project uses TypeScript or JavaScript
• ESM (import) or CommonJS (require) style

Step 4.2: Confirm with the user

Briefly share what you found and where you plan to put the new files. Ask if this looks right.

Step 4.3: Generate the helper module

Create a helper module that wraps HTMLtoDOCX with sensible defaults. Place it where the project keeps its utilities (e.g., src/lib/, src/utils/, lib/).

The helper should:

• Import HTMLtoDOCX using the project's import style (ESM or CJS)
• Export a function like generateDocx(html, options?) that calls HTMLtoDOCX and returns a Buffer
• Include commonly useful defaults (font, margins) that the user can override
• Use TypeScript if the project uses TypeScript

Step 4.4: Generate integration code

Based on what the project needs:

If it's a web server (Express, Fastify, NestJS, Next.js, etc.): Create an endpoint that accepts HTML (in the request body) and returns a .docx file. Match the framework's routing pattern:

• Express: Router with res.setHeader('Content-Type', 'application/vnd.openxmlformats-officedocument.wordprocessingml.document') and res.send(buffer)
• Next.js App Router: app/api/.../route.ts exporting POST function, return new Response(buffer, { headers })
• NestJS: @Controller + @Post with @Res() res for streaming the buffer
• Other frameworks: match their conventions

Set the response headers:

Content-Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document
Content-Disposition: attachment; filename="document.docx"

Wire the route into the app's main file (unless file-based routing like Next.js).

If it's a standalone/script project: Create a script that reads HTML (from a file, stdin, or a string) and writes a .docx file. Include a basic example HTML string to demonstrate.

Phase 5: Verify and Summarize

If the project uses TypeScript, run npx tsc --noEmit to check for type errors.

Print a summary:

• Files created and modified
• Package installed
• How to use (example curl command for endpoints, or how to run the script)
• Link to docs: https://docs.turbodocx.com/docs
• Key options they can customize (page size, orientation, margins, fonts, headers/footers)