Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Help getting started #57

Open
charliebrownfaz opened this issue Dec 3, 2024 · 5 comments
Open

Help getting started #57

charliebrownfaz opened this issue Dec 3, 2024 · 5 comments

Comments

@charliebrownfaz
Copy link

I've been trying to setup Smart Connections software with Obsidian for the past few days.

I've installed the plugin, the Smart Connection software on my previous vault, deleted and re-installed a fresh copy of Obsidian and am able to access the Smart Chats feature.

I'm trying to create a CustomGPT to query my Vault, but I cannot make the connection work. Whenever I test Run Tests for custom-1 I get "Error running test custom-1/alignment "handles_missing_alignment_file". HTTP error! status 404" as an example.

I've tried to follow the video instructions closely, but am getting frustrated at my inability to get it working. Any assistance most gratefully received.

@brianpetro brianpetro transferred this issue from brianpetro/obsidian-smart-connections Dec 3, 2024
@brianpetro
Copy link
Owner

@charliebrownfaz it seems you are trying to use the Smart Connect app to connect your vault to ChatGPT, so I moved the issue to the relevant repo.

Did you install the Smart Connect software and configure your connection key?

https://github.com/brianpetro/smart-connect?tab=readme-ov-file#install

https://docs.smartconnections.app/Smart-Connect-app/

🌴

@brianpetro brianpetro changed the title Smart Connect never connects - Is this feature depreciated? Help getting started Dec 3, 2024
@charliebrownfaz
Copy link
Author

Hi Brian, thanks for taking the time to reply, sorry that I put my issue in the wrong section.

I have followed your video down to the second, I am able to find the http://servername.loca.lt/openapi.json however, when I try to import this into ChatGPT, I get nothing, this point in your video where your json file is imported into the section below for the schema.

I have tried to simply copy the file into the schema section, but I then get the 200 error

I don't have a Supporters License key, do I need a Smart Connect Key?
CleanShot 2024-12-03 at 15 49 43

@brianpetro
Copy link
Owner

So I can't do much to help with running your own tunnel server since there are so many variables.

At a brief glance, it looks like your server URL is misconfigured.

However, I can tell you that if you're trying to use a free tunnel service it probably won't work because the free services require inputting a tunnel password, for which there is not built-in way to do via the ChatGPT to Smart Connect connection. My understanding is that this can be bypassed if you use the premium version of those same services.

Lastly, you don't need to be a supporter to use Smart Connect. A connection key to use the official tunnel service can be purchased here https://smartconnections.app/smart-connect/

🌴

@charliebrownfaz
Copy link
Author

Thanks Brian, that was sufficient information for me to be able to problem solve the issue which has stumped me for days.

I have been able to set up a reverse-proxy through my Synology NAS, pointing to my computer running Smart Connections and Obsidian. In case it might help someone else, here is a screenshot of my Synology Settings.
CleanShot 2024-12-03 at 18 20 31

The Plugin is working very well in general, although it does think some of my files are empty, and I am still getting this error below on setting up the Schema, but it working sufficiently for me to be able to start to evaluate it's usefulness.
CleanShot 2024-12-03 at 18 18 42
Thank you once again for your assistance and for taking the time to create the plugin.

@brianpetro
Copy link
Owner

Thanks for sharing how you are creating your own tunnel connection 😊

I was able to track down the issue you're having.

In the custom GPTs section, manually add the available actions (see screenshot)

Screenshot 2024-12-03 at 2 22 28 PM

When no actions are explicitly added, "all actions" are added. And there are apparently a couple bugs in how "all actions" are added, 1) components aren't being included, and 2) actions that shouldn't be included are being included.

I made a note to get those issues fixed before the next release (should be this week).

In the meantime, manually adding those actions should get things working for you 🌴

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants