Why I Built VertiWiki: A Zero-Backend, Single-File Markdown Wiki Engine

Octadira — Sun, 30 Aug 2026 09:17:52 +0000

When building project documentation, knowledge bases, or internal wikis, developers typically face a choice between two paradigms:

Static Site Generators (SSGs) like Docusaurus, VitePress, or MkDocs. These generate great static HTML, but require a local Node/Python runtime and a CI/CD build step every time a markdown file is updated.
Database-backed CMS Wikis like Wiki.js or BookStack. These offer instant editing, but require databases, backend daemons, and ongoing server maintenance.

I wanted a third option: the simplicity of raw Markdown files on disk, combined with instant client-side rendering and zero backend requirements.

That led to VertiWiki.

The Architecture: Single-File Distribution

VertiWiki is distributed as a single standalone HTML file (vertiwiki.html or index.html).

Using Vite 6 and vite-plugin-singlefile, all application logic, routing, stylesheets, and third-party libraries are bundled directly into that one file:

Markdown Parser: marked
Security: DOMPurify (sanitizing every HTML output before injection)
Syntax Highlighting: prismjs
Math & Diagrams: katex & mermaid
Search Engine: minisearch (in-memory client-side fuzzy search)

Example Directory Structure

.
├── index.html          <-- Standalone VertiWiki bundle
├── config.json         <-- Wiki title, theme, analytics config
├── navigation.md       <-- Sidebar navigation hierarchy
├── index.md            <-- Home page
└── docs/               <-- Your raw markdown files

Zero Build Step for Content

Because the engine runs in the browser, you never need to run npm run build when authoring or updating content:

Edit or add any .md file in your repository or server folder.
Refresh the browser.
The client fetches the raw Markdown via standard fetch(), executes the plugin pipeline, sanitizes the HTML, and renders it dynamically on the fly.

Client-Side Offline Search

Traditional static sites either build massive static search index JSON files at compile time or rely on third-party search APIs (like Algolia).

VertiWiki indexes documents dynamically into MiniSearch on the client. Pressing ⌘K or / opens an instant fuzzy-search modal that scans article titles and body content with zero network calls.

Open Source & Getting Started

VertiWiki is 100% open source under the MIT License.

GitHub Repository: github.com/Octadira/vertiwiki
Live Demo & Docs: verti.wiki

You can grab the latest vertiwiki.html from the GitHub Releases, drop it into your folder of markdown files, and host it anywhere (Nginx, Caddy, Cloudflare Pages, GitHub Pages, or an S3 bucket).