Appflowy - open-source Notion Alternative: how I tested it with Web2MD

Appflowy is one of the more serious open-source Notion alternatives I have tested. It is not a perfect Notion clone, and that is probably a good thing. The pitch is different: local-first workspaces, open-source code, self-hosting options, and more control over where your notes, docs, and project data live.

I tested Appflowy with the same workflow I use for most research-heavy writing: collect a few pages, pull the useful parts into Markdown, then send that Markdown into ChatGPT, Claude, or Cursor for summarizing, rewriting, or coding work.

That last step is where Web2MD fits.

Web2MD is a Chrome extension that converts web pages into clean Markdown for AI tools. It runs inside your browser, so it can read pages you are already logged into, including private workspaces, internal docs, dashboards, and some paywalled pages. Server-side readers usually cannot reach those pages because they do not have your browser session.

For Appflowy, that browser-side detail matters. Open-source tools often have docs spread across marketing pages, GitHub, guides, changelogs, and logged-in product screens. If you are comparing Appflowy with Notion, Anytype, Obsidian, or Outline, you do not want to paste messy page text into an AI tool and hope for the best. You want structured Markdown.

What Appflowy is good at

Appflowy is an open-source workspace app for notes, docs, tasks, and structured data. If you like the Notion style of blocks and databases, Appflowy will feel familiar. The difference is mostly philosophical and architectural.

Notion is polished and hosted. Appflowy is more transparent and gives you more control. You can inspect the code, follow development, and choose workflows that fit privacy-sensitive teams better than a fully hosted SaaS product.

When I tested it, the parts that stood out were:

• The interface is close enough to Notion that most people will not feel lost
• Pages and databases cover the usual team knowledge base use cases
• The open-source model makes it easier to trust and audit
• Local and self-hosted direction matters for teams with privacy rules
• The product still feels younger than Notion in polish and ecosystem depth

That last point is important. If you need the deepest template ecosystem, the most integrations, and the smoothest mobile experience today, Notion is still hard to beat. Appflowy is more interesting if you care about ownership, portability, and avoiding lock-in.

Why Markdown matters when evaluating Appflowy

AI tools work better with clean input. That sounds obvious until you paste a modern web page into a chat window and get navigation links, cookie text, sidebar labels, duplicated headings, and half the footer mixed into your prompt.

I used Web2MD to convert Appflowy pages into Markdown before asking Claude and ChatGPT to compare features, summarize setup steps, and turn documentation into checklists.

Here is the kind of cleaned-up Markdown structure you want from an Appflowy product page or doc page:

# Appflowy

Appflowy is an open-source workspace for notes, tasks, and project documentation.

## Main features

- Document pages with block editing
- Databases for structured information
- Local-first direction
- Open-source codebase
- Options for teams that care about data ownership

## Best fit

Appflowy works well for users who want a Notion-like workspace but prefer more control over their data and tooling.

That is much easier to use in an AI prompt than raw copied page content. You can ask:

# Task

Compare Appflowy with Notion for a 10-person engineering team.

## Source notes

- Appflowy is open source
- Notion has a larger template and integration ecosystem
- Appflowy is better for teams that care about local control
- Notion is more polished for nontechnical users

## Output

Write a balanced recommendation with tradeoffs, not a sales pitch.

The difference is small on one page. It gets large when you are collecting ten or twenty sources.

Where Web2MD beats server-side readers

Tools like Jina Reader and Firecrawl are useful. I use server-side readers when I need a quick public-page extraction, crawling, or a workflow that runs outside my browser. Firecrawl is especially strong when you need developer-facing scraping and crawling infrastructure. Jina Reader is convenient because you can often prepend a URL and get readable text quickly.

MarkDownload is also worth mentioning. It is a good browser extension for saving pages as Markdown, and it has been around for a while.

Web2MD is built for a slightly different job: getting the current browser page into clean Markdown for AI tools with less friction.

The browser-side model is the main advantage. If you can see the page in Chrome, Web2MD can usually convert it. That includes logged-in documentation, private dashboards, internal wiki pages, and account-specific pages that a hosted reader cannot access.

That matters for Appflowy research because a real evaluation is not just the public homepage. You may be looking at your own Appflowy workspace, imported docs, team notes, or private project pages. A server-side tool cannot read those unless you expose credentials or build an integration. I would rather not do that for routine AI prompting.

Web2MD also keeps the workflow local and private. The conversion happens in your browser. You do not need an API key for the free tier. You get 3 conversions per day for free, and Pro is 9 dollars per month if you need more. That limit is real, and it is worth saying plainly. If you need hundreds of conversions per day or automated crawling, Web2MD is not trying to replace Firecrawl. It is better for one-page-at-a-time research, writing, coding, and AI context capture.

The token counter is not a small feature

The built-in token counter turned out to be more useful than I expected.

When you are sending Appflowy docs or comparison notes into ChatGPT, Claude, or Cursor, the question is not just "did I copy the page?" It is "how much context am I about to spend?"

A long documentation page can look reasonable in the browser and still eat a large chunk of your context window. Web2MD shows the token count before you send the content onward, so you can trim the source or split it into chunks.

For example, I converted an Appflowy-related page, checked the Markdown, removed repeated navigation, then sent only the setup notes into Claude. That gave me a tighter summary and fewer generic answers.

This is the kind of boring feature that saves time. It does not sound flashy, but it prevents bad prompts.

How I would use Appflowy and Web2MD together

If I were evaluating Appflowy for a team, I would use this workflow:

1. Open Appflowy pages, docs, GitHub issues, and comparison pages in Chrome
2. Use Web2MD to convert each useful page to Markdown
3. Watch the token count before sending anything to an AI tool
4. Send the Markdown to ChatGPT or Claude for summaries and tradeoff analysis
5. Use Cursor if the source includes setup docs, API notes, or code-related instructions
6. Save the final Markdown notes back into Appflowy or another knowledge base

That loop works because Markdown is portable. You are not trapped in one AI chat, one workspace app, or one export format.

Limits I noticed

Web2MD is Chrome-only right now. If you live in Safari or Firefox, that is a real constraint.

The free tier gives you 3 conversions per day. That is enough for testing, occasional research, and light AI prompting. It is not enough for heavy research days. Pro is 9 dollars per month, which is reasonable if you use it regularly, but it is still another subscription.

And Web2MD is not a crawler. If your goal is to scrape an entire documentation site, Firecrawl is a better fit. If your goal is to convert the page you are currently reading, especially a private or logged-in page, Web2MD is the cleaner option.

Appflowy verdict

Appflowy is worth trying if you want an open-source Notion alternative and you care about control. It is especially interesting for technical teams, privacy-conscious users, and people who dislike having their entire knowledge base locked inside one hosted product.

It is not the obvious winner for every Notion user. Notion is more mature. Appflowy is still catching up in polish, integrations, and ecosystem depth. But Appflowy has a clear reason to exist, and that makes it more compelling than yet another notes app with a database view.

For AI-assisted research, Web2MD makes the evaluation easier. It turns Appflowy pages, docs, and private browser content into Markdown that ChatGPT, Claude, and Cursor can actually use. It also keeps the process local, shows the token cost, and avoids API setup.

If you are comparing Appflowy with Notion, start by collecting clean source material. Try Web2MD on the public Appflowy docs, then test it on your own workspace pages. You can read more about browser-side conversion in our guide to turning web pages into Markdown for AI tools, or install Web2MD from web2md.org and use the free 3-conversion daily tier before deciding whether Pro makes sense.