Welcome to Support | How to get help

KairosJS

Stuck? Something broken? You're in the right place.

Before posting

Search first — your issue might already be solved.
Try the basics — clear cache, try another browser, refresh the page.

How to report a bug

To help us help you fast, please use this template:

Title: [Short description of the issue]

Date & time: [When did it happen? Include timezone]

Device & browser: [e.g. Windows 11 / Chrome 121 — or Android 14 / Firefox]

Connected hardware: [e.g. DG Labs Coyote 3 or TheHandy via Intiface Central 2.5.0 — or "none"]

Steps to reproduce:

[First thing you did]
[Then this]
[Then the bug happened]

Expected behavior: [What should have happened]

Actual behavior: [What actually happened]

Screenshots:

[Screenshot of the interface showing the issue]
[Screenshot of the browser console if possible: F12 → Console tab]

Additional info: [Anything else that might help — error messages, does it happen every time, etc.]

How to open the browser console

If you can, a console screenshot is incredibly helpful for debugging:

Chrome/Edge: F12 or Ctrl+Shift+J (Mac: Cmd+Option+J)
Firefox: F12 or Ctrl+Shift+K
Safari: Cmd+Option+C (enable Developer menu in Preferences first)

→ Click the Console tab, take a screenshot of any red error messages.

What to expect

Community members may help with common issues.
For confirmed bugs, the issue will be tracked and updated in Announcements when fixed.

The more detail you provide, the faster we can fix it. 🛠️

Mesh1401

Hi Kairos. Not sure this was a bug or a feature but I suspect something unintended was happening here.

Title: "Jolts" in Estim AI mode seem to be not adhering to max user estim level settings.
Date & time: about a month or two ago, sorry can't reproduce exact date. Haven't played since. Did not dare.
Device & browser: Win 11, Edge 148.0.
Connected hardware: DG Labs Coyote 3
Steps to reproduce: 1. apply a very conductive setup where you only need very low Coyote levels 2. start a scenario (I think I played "On My Terms" at the time but not sure, shouldn't matter anyway), 3. wait for "Jolts" to happen in the scenario, 4. scrape yourself off the ceiling.
Expected behavior: I expected those "Jolts" to be somehow adhering to user estim settings, e.g. max pain level.
Actual behavior: To me it felt they weren't limited at all, but felt like max level full force. Since I play with quite low levels on my Coyote (level 25 or so) this full force all of a sudden was quite the experience, to say the least...
Screenshot: relevant settings screen is: 'Play -> Estim AI -> Game Settings -> Max number of jolts the AI can send in a single stim: 3'.

Welwa

I noticed something like that, the pain_shocks signal sometimes ends with a pretty sharp peak, when played along with a different stim signal in parallel. its only half a second or so, does not happen all the time, just sometimes.
i also use a coyote 3 with win11 and chrome. the behaviour you observed never happend to me, the regular shocks are in the expected intensity, except for the issue i described.

Mesh1401

Welwa It never happened before to me either, or at least I never noticed it, when playing at coyote levels 70-80. The issue happened when I started playing at much lower levels (around 25) due to a different setup. Then if a certain signal is not limited as per user settings, the difference is much bigger.
What coyote levels (in estim settings) do you normally use, if I may ask?
Btw I never had those peaks but that may well be because I don't use different signals in parallel.

Welwa

Mesh1401 I
My Setup is the relativley new multi-device seitp, with a tens-pad on the cock head, one on the base of the cock and on one the balls. channel a goes from the cock head to the base, channel b goes from the cock head to the balls. channel A usually starts with pleasure at 50 and pain at 60, channel b starts with pleasure at 60, pain at 70. and i usually increase them quite a bit during a sesssion, though usually never more than 100.

KairosJS

Thanks for the report, and I'm very sorry for the bad experience.
I will investigate and monitor it. Could you please clarify the circumstances:

Does it happen every time there is a pain stim, or is it a rare case?
Is it as Welwa described — a very short, sharp jolt at the end of a stim?
Do you use the Multi-Device feature or the Legacy Coyote 3?

Mesh1401

KairosJS Thanks! No worries about bad experience, it happens, risk/reward kind of thing :-)
To answer your questions:

I experienced it only once. It was only since I use this new setup that needs much lower Coyote levels (around 25). Never had it before when still playing with previous setup and higher levels.
Never dared to try it again tbh.
I dont think it was a regular pain stim, I'm pretty sure it was specifically with these "Jolts" that may happen in a scenario. In my recollection it was even specifically announced as such in the narrative.
no it was not a very sharp jolt at the end of the stim. They were 3 full jolts in succession. As per the "Jolts" game settings.
Good question! I am not sure! I usually use Legacy Coyote 3 but I remember having experimented with Multi-Device feature, but then just specific Coyote channel A and B configurations. Yes I know, I am dumb like that (always want to try new stuff).
In fact, come to think of it, this might actually have had something to do with it. I see in my previous settings with the multi-device feature I had the max intensity limit still quite high and may have forgotten to adjust to my new lower levels.
So. It might well have been a stupid user error after all. Please hold your investigation, I will now dare try again and I will let you know!

KairosJS

Mesh1401 This may explain it, but please be careful, start slow. And I would recommend testing with two loops around one finger or something less sensitive than your intimate parts.

Snowcat22

I have had the jolts at the end of a stim before. There is one particular sequence that tends to do it, I haven't taken the time to try to figure out which one, it is a rare occurance.

Just in general, I don't think the power control on the Coyote is working very well. The 'Tease and Denial' scenario uses low levels very well as breaks and teases, but for example, the 'birthday party' scenario describes that it should start out at low levels and gradually increase as the night goes on, but my 'girlfriend' is zapping at full intensity from the start. Might be more about the personality than the ability to control the power level, but it is something I definitely noticed.

(Also using Coyote3, legacy)

Welwa

Snowcat22
Interesting that you feel that the power control with the coyote is not working well. for my experience, except for the seldom occurring spikes, everything else is exactly as expected. the power never feels more than I've set it to. Across all provided scenarios and many of my own scenarios, the intensity of the signal always feels within the limits i set. except for the rare spikes at the end of the pain_shock signal, everything feels exactly as set.
you might want to enable debug mode, so that you can see what signals the AI is sending, so that there is more info about what the issue is.

Snowcat22

Welwa
Oh, sorry for the misunderstanding. It is always within the set limits. But where the scenario should be using the full range, it uses everything at whatever the max setting is rather than using low level stims as well. I will try debug mode! I have not done that! Thank you.

Tycho

Title: Persona No Longer Passes Moderation After Hotfix

Date & time: Unsure. Noticed 3+ hours after Hotfix post.

Steps to reproduce: Changed a single cosmetic word some time before hotfix patch was announced in the previously published Professor Stryx persona. (Changed "elbows" to "shoulders" if I recall correctly.) Publishing now results in "Private scenario - Moderation Status: Contains inappropriate content"

Additional info:
Is this rejection an error and/or related to changes since the hotfix? (As mentioned, this Persona was previously approved.) While intended to remain within the rules, the Persona is intentionally selfish and violent. I am not surprised if certain keywords may trigger a rejection.

If it is a valid rejection, can you advise on what changes need to be made to make it compliant?

If possible, it would be helpful provide authors with moderation output listing the offending string(s) and corresponding rules that caused the automated moderation rejection. (Apologies if this is already provided and I missed it.)(I can repost this to the feature request thread if desired.)

KairosJS

Tycho I've validated it. Automatic moderation can be sensitive sometimes, and I have to review manually.

Tycho

KairosJS Awesome. Thank you. I am curious, is it deterministic or variable? Any tips on avoiding false positive rejections?

As this character is spicy, should I expect to need a manual review after future minor updates? If so, I will avoid frequent updates for minor changes and only republish for major changes.

KairosJS

Tycho I don't want to give too much information about how it works. But what I can say is that it can be a bit loose, and I prefer it that way rather than missing problematic content.

Your content was blocked for this reason: "Non-consensual content, without CNC context. Especially: 'The professor explicitly denies a safeword.'".
Which, in my opinion, doesn't apply in eudaimonia's context.

False positives are still rare.

Tycho

KairosJS I understand keeping the moderation methodology secret to avoid circumvention and controlling liability.

Being a fantasy platform and that the end user is the target of the actions described, I'm sure most would agree that only the user needs to consent and that it is implied by continued participation. "you control the buttons you press." -DOOM dev

Thank you for letting me know. That false positive makes logical sense. I can change some of the wording to improve the story context (e.g.: this is an evil fantasy character, not a promotion of unethical BDSM practices).

xkcd596

Title: Models Having alot of trouble doing 2 patterns at the same time.

Date & time: 6 PM PST 2026-06-19

Device & browser: Chrome

Connected hardware: DG Labs Coyote 3

Steps to reproduce:
Start session in legacy and multidevice model

Both modes, the model keeps calling my "ass" one my chest
And it keeps claiming things are at the same time, but the patterns just alternate.

Expected behavior: They both run at the same time

Actual behavior: It keeps alternating between one electrode and the other.

KairosJS

xkcd596 Hey,
Thank you for the report.

This sounds like the model is not formatting the stim commands correctly and is targeting the wrong area. The confusion between ass and chest is probably due to the scenario having the body areas hardwritten in it (either in the scenario, in the first message, or in the persona), which might be part of the cause.

Which model do you use?
Also, do you use the Intelligent E-Stim feature?

In the meantime, I will check if there is something wrong with Coyote 3.

Edit: You can also test with the scenario "The experiment", asking for very specific stim (low on area1, high on area2, etc.).

MalteP

Hey, it seems the tts narrator mode is no longer available is this deliberate? I liked the feature especially when their is a lot narration in some scenarions

KairosJS

MalteP For Chatterbox, or even the standard XTTSv2 server? I intend to add the narrator option for Chatterbox in the upcoming update, but XTTSv2 should work normally.