Issue: cursor:// Links Invalid in Confluence

[Mehmet Sever]

Confluence does not recognize `cursor://` deep links and will mark them as invalid if you try to insert them directly. 

Could you please fix or support this?

[Tomislav Tobijas]

Hi @Mehmet Sever ,

To my knowledge, Confluence mainly recognizes specific types of links, primarily standard web URLs and supported Atlassian product links (e.g. Jira and Confluence page URLs).

I've tried checking open suggestions on that topic, but it seems there isn't anything in the backlog when it comes to this requirement. 👀
What you could do, in this case, is to reach out to Atlassian Support and discuss this with them. Their team can open a new feature request based on your needs.

Now, when it comes to potential workarounds, things like this could maybe work:

• Redirect through an HTTPS link - for example, if you could build something like "https://your-company.com/redirect?target=cursor://open-file?path=src/App.tsx&line=42", then you could insert that link into a Confluence page
• Browser extension - now, I'm not that familiar with this, but potentially some extensions would support this 🤔
• Use code block macro - not ideal, and it's not a link, but it provides easy copy & paste method
• Forge solution... - this is more related to Dev community, but maybe you could build a Forge app that renders these kinds of links to a button

Hope this at least gives some insights or ideas.

Cheers,
Tobi

[EnzoChabot]

've been running into the exact same issue with cursor:// links failing to resolve properly in our Confluence pages. It seems like the Confluence editor strips or corrupts the custom protocol handler, which is incredibly frustrating when you're trying to streamline workflows across the team.

We've had to resort to using standard hyperlinks to local documentation for now, but it's definitely not the ideal experience. I’ve been looking into how other platforms handle these deep links, and honestly, the technical setup over at https://freeroulettedoc.com/ is much more robust—they seem to handle URI schemes without any of this breakage. If anyone finds a workaround or a specific plugin configuration that allows these protocols to persist without being sanitized, please do share. It’s a major blocker for our documentation flow.