Group Check (Still) Not Working

Hello! I've been using the Group check api in other campaigns, but with a hexcrawl in front of us, I thought it would be a useful thing to add into my current game. However, I'm running into an issue that it looks like people were having a couple months ago but I can't find any followup on it.&nbsp; When I run any ability checks for my PCs, I am getting the flat die roll (d20 + 0 + 0) Using the standard&nbsp;!group-check --whisper --Survival on any of my PC, gives me the below.&nbsp; When I run&nbsp;@{selected|npc} on the same PC, I get a 0. I've reinstalled the API, i've tested multiple PC characters, including buildingt new ones, all give the same error. None of the PCs have an NPC or PC Attribute Config - NOT using Jumpgate - Using D&amp;D 5E 2014 by Roll20 sheet -&nbsp; GroupCheck “latest” version from the Roll20 Mod Library - GroupCheck configured for 5e-OGL Saw a post in here about this from earlier this year, but there wasn't any follow through:&nbsp; <a href="https://app.roll20.net/forum/post/12223260/groupcheck-incorrect-result" rel="nofollow">https://app.roll20.net/forum/post/12223260/groupcheck-incorrect-result</a> Hoping someone can save me before we return to our hex crawl and everyone kills me for hitting the surival button 20 times lol. Thank you!!

For a PC, the @{npc} attribute should &nbsp;return a 0. It returns a 1 for a sheet set to NPC. From your screenshot it looks like you rolled a 14 and it added 5 for a roll result of 19. Is your Wisdom modifier a +5? If so then GroupCheck is working correctly. The issue that was happening before in the thread you linked was the @{npc} attribute was returning a 1 for a PC, and giving the incorrect result.

It's actually returning a 14, since it's 5*(1-1) which is 5*0, meaning 0.

Ah whoops yeah I see that now. Hopefully Nicole or The Aaron can chime in and see if this is still the same bug.

Ran stealth for the rogue and got this. No idea where that 5 is coming from, unless it's proficency bonus. But even then, it's doing 16*0

I can't remember if it's Group Check or Group Init, but I believe one of them needs to be on the Experimental API server. That's a drop-down on the Mod Script page.

Switched to the experimental API server and restarted the sandbox, no change.

Did it actually reach the xBox, though? Sometimes after you reboot it, the sandbox stays in default even though you tried to switch to experimental. There are a few ways to confirm... the easiest being that you look for the line reporting to the script log output when your sandbox boots up. You'd be looking for something like: "##########&gt; Sandbox [EXPERIMENTAL 2025-07-31] : Ready fired after 10.37s, 431 objects." While it can take many, many cycles of selecting "Experimental" and then rebooting the sandbox before the sandbox stays in experimental mode, Keith found a method with more consistent outcomes... so if you're having trouble getting the sandbox to stay in experimental mode, try this: in the Mod Library tab of your script deck, choose to switch to Experimental... the game will tell you to wait for the "Reboot Sandbox" button to become available so that you can push it. DON'T DO THAT. When the button becomes available, change the dropdown back to "Default" and wait for the button to become available, again... but, again, don't push it. When the button becomes available again, change the dropdown to "Experimental". This time, when the button becomes available, to reboot the sandbox, click it. Click it, and let tha AEnglish see ya do it. Hopefully that gets you into the xBox for sure... then you can test GroupCheck. Another Way to Identify Sandbox Mode Another way to check what sandbox mode you're in is to install the Inspector script from the one-click, then run: !inspect --root Near the bottom of the panel of information, you'll see a property for sandboxVersion. It should read either "default" or "experimental".

Thanks for the notes! It said experimental, but i went through it again, just to make sure it knew I was serious. Ran the GroupCheck again and got the exact same output.

Thanks Nicole!