The Notion crawler

When someone pastes a URL into a Notion page and clicks “Create bookmark,” Notion’s server-side crawler fetches the page and extracts meta tags. The user agent contains “Notion” and identifies as a bot. No JavaScript execution.

No JavaScript execution

Notion’s crawler doesn’t run JavaScript. OG tags must be in the initial HTML response. SPAs that render meta tags client-side need SSR or pre-rendering.

Response requirements

Expects 200 OK with text/html. Redirects are followed, but excessive redirect chains or auth walls cause failures. The crawler also reads the page’s favicon for the bookmark card.

Bookmark blocks vs. Link Preview integrations

Notion has two URL features that people confuse:

• Bookmark blocks – the standard OG-tag-based preview described here. Available for any URL.
• Link Preview integrations – API-powered previews for approved partners (GitHub, Figma, Jira). Richer, interactive content. Separate system from OG tags.

Checking if Notion can reach your page

curl -A "Notion" https://example.com/your-page

If your <meta> tags are in the response, Notion will see them.