Tried using two - windows - too cumbersome - not worth it to me
Would rather have AI built in
Tried to go to License - no upgrade path…
Hi @MichaelDroste,
Nice, the Writing Assistant and the MCP really are worth having! If you get into using the MCP I think you’ll be blown away by how cool and useful it is. We also have some more tutorials in the works to show you the kinds of things you can do with it. Stay tuned for those.
Anyway, there’s no upgrade path (yet) so for the moment we have to do it manually. i.e. we’ll issue you a refund, then you can buy the Studio license.
Let me know if you want to go ahead with that, and I’ll refund your order!
yes please
Yes, click that button!
I have also gone ahead and issued you with a full refund, so you’re dafe to go ahead and purchase the Studio License ![]()
Thanks for your support!
@dan Took a little while - Up and running with Gemini - needed to use 2.5 Flash
The text writing works!!!
How do I do tags, categories, seo description, and titles?
Tried highlighting, selecting etc…
Not working yet?
No, the writing assistant will only work on the main body text.
For anything else you should use the MCP, have you watched this video?
FYI: As far as I know the Gemini App doesn’t yet support MCP servers. On the Mac you can use Claude or Codex (OpenAI/ChatGPT) to connect to your blog via MCP.
What a pain:
Ah, gotcha. If you just locked in a year of Gemini Advanced (the consumer subscription through Google One), that changes the calculus.
Here is the realistic, unvarnished breakdown of how Gemini handles MCP right now, including the limitations, the positives, and your alternative options.
If you are chatting directly inside the standard Gemini web interface (or the mobile app), you cannot plug in custom, third-party MCP servers. Unlike Claude Desktop or Cursor—which let you paste in an MCP configuration file to talk to any local folder or external tool—Google keeps the consumer web app closed to outside protocols.
Instead of MCP, the web app relies entirely on Google’s proprietary Extensions.
Even though the main web app is closed to the protocol, Google has leaned heavily into MCP for developer environments and local setups. If you want to leverage Gemini alongside MCP servers, you have a few ways to do it:
Google provides an official, open-source command-line tool called gemini-cli.
If you want a visual interface where Gemini uses MCP to look at local files or data structures, you can use your Gemini model access inside third-party apps like Cursor or VS Code (using extensions like Continue).
If your primary goal in wanting MCP was simply to get Gemini to interact with your own personal files, documentation, notes, or spreadsheets, you don’t actually need to set up an MCP server.
Because you have the paid subscription, you can turn on the Google Workspace Extension directly in your Gemini app settings. This allows the web interface to natively read, analyze, and pull live data from your Google Drive, Gmail, and Google Docs using simple @ mentions right in your chat.
I thought Google Gemini ‘did it all’ signed up for a years subscription…
Basically for elements - useless 2 paragraphs at a time - doesn’t cut it for me…
I’ll go back to manually copying in the meta information from the previous post
Go back to creating in the ai
Etc…
For me - kind of a bust… I should’ve gone with Claude…
You could always contact Google for a refund…
I’ve found it’s best to go monthly with the AI subscriptions as I often want to change and try out other services. The platforms are moving really fast so you don’t always get the benefits of the new features if you stick with just one.
Trying to cancel - Google can’t cancel as bought through Apple Pay
Apple Pay says a decision by May 28…
Hopefully
it will be cancelled!