#development
1 messages Β· Page 15 of 1
i think you basically have no understanding of what it is you are talking about
Well, you seem to be an expert in the field, we'll wait for your PR to fix the problems π
I think you have zero information to reach that conclusion
drop a Make no mistakes too
I know you're new to this but it sounds like you don't much about it
to clarify, the pr you linked provides an AGENTS.md file for use by a developer's AI agent when they develop a plugin. it would be used to provide context when a developer asks "make me a plugin that does X"
Yeah that's attempting to be a skills file but the whole file is structured to be an adversarial review agent
ok, write us a better one
what you're saying makes sense on a surface level and we even talked about that a little bit internally, but empirically a handful of us tested it and we got good results on plugins so idk
I would gladly help out I'd appreciate if you were a bit nicer to me
That makes sense, from a technical perspective you liked the output, is that fair to say or was there a different heuristic you judged it by?
problem with a skill approach is you're rolling the die whether the agent decides it's relevant to read the file
more or less, it was very informally tested (at least by me i am sure moses hammered it a lot more)
That's handled at the harness level
I think the file is always read into the context right
it depends on the tool
well yea
with skills, just the description at the top is guaranteed to be in context
agents.md is handled separately
moses is probably the best person to ask about this
Claude thinks it's special π
Skills files aren't always fed into context but they can be programmatically read with a hook
i dont want to write hooks for every model provider
In theory I think the file works great as a review. Before sending a pr, do an adversarial review of the code that's been generated against these standards refractor when directed etc.
it might be good if we went and found all the vibecoded submissions (so you know all of them) and found the ones that were doing stuff we don't really like (net http, thread sleeps etc) and left a comment that said "hey please pull this into your agent and make any changes it recommends"
i think we already did that for one or two submissions
Yeah I think that's a good approach.
I also wanted to say I think this is an awesome step to handle this problem, which all FOSS are currently struggling with. I don't mean to detract from the work that's been done
Something that may be a valuable add is forcing this review before a submission even hits prs by setting a blank review.md file and have that agent write to review.md (intended to be called before submitting pr) if it passes. If the review.md is blank the pr is put in waiting for author state instructing the user to run the agentic review on their. Human submittions just enter "I am human" in the review.md file
wouldnt that mean everyone submitting prs would have to have an agent review it first?
There's room for improvement and not a foolproof method by any stretch
No, humans just type I am human
i dont want to do that
Any reason?
i dont want to force people to do that
Maybe there's a simpler verification loop that I'm not aware of
the purpose of this whole agents.md is not for our security or for our review of plugins directly, it is to reduce the amount of bad code submitted to us so that when we do our actual review there are less problems
more specifically, what you are proposing here is degrading the human pr submission process in exchange for maybe getting less bad automatically generated AI submissions, which is just a tradeoff im super uninterested in doing really and seems like a bad idea
I get that, and it makes sense to not raise that bar. I did jump the gun on the intent of this but it became clear later on. I think there is a clean solution with some more iteration. generally 100% the right line of thinking here though
When items are in the inventory they get a little drop shadow, anyone know what widget setting controls that? I have bordertype = 1 for the nice black outline but wanted that drop shadow too
no api exposed to set it
ahh damn, ill just have to draw my own
d54603a gpu: disable priority sort for streamed transpa... - Adam-
7132534 gpu: disable priority sort for streamed transpa... - Adam-
for real this time
ce5c433 update demonic-leagues-task-checklist - alexshiao99
bd87403 update traducao-osrs-br - walterrezende12-tech
didn't take long before someone managed to get the bot to merge something it shouldn't have lol
i just forgot about that totally
i didn't even know about it
its also only 6mb
i'm guessing their ai recommended they do it so they can embed more static translation data
they added a 16mb json to the repo, but ofc that gets compressed when put into the jar
oh, not sure if the thing we have limits the uncompressed or compressed size
i'd assume it's just the jar itself
but in that case, not sure why they added that property or even knew to look for it due to having a large file lol
without even knowing the size of the jar
its all just ai
guess that'd make sense
The positive prompts will be what youβre actually prompting it to do. Do build me a plugin that tracks data about flipping but totally different this time
My AI suggested adding that setting because of the larger translations file, but I understand now that it was not needed. Sorry if I did something wrong there.
I opened a new PR with the correction, without that setting, and with the updated plugin source aligned properly.
Does the AI reviewer do changes to existing plugins as well or just new ones?
yes
yours isnt checked automatically right now because the ai cant be trusted with it
1a7f76b update prayer-time-infobox - arbeerby-pixel
10727cb update traducao-osrs-br - walterrezende12-tech
9c3cbcc fix hub plugin missing README not showing 'Repo... - tylerwgrass
33d2c7e update inventory-restore-values - Sintry1
763cf92 Deploying to gh-pages from @ runelite/runelite.... - Adam-
It's ok?
i think so, is there a reason it wouldnt be?
I was just worried there might have been another issue with the automated approval, since the previous PR was reverted right after merge. That's all. Thanks.
Is the PR backlog longer than normal these days
What the zuck
One sec, we're celebrating having dropped to only 168 open PRs
we need an updated graph
The current one almost warrants a pin
Are that many plugin developers really using it
I ain't heard of Codex until just now
Yeah, I would have figured it would have spiked earlier with claude code etc. But I guess it really got bad when it wasnt a CLI
Reminds me of a quote by a wise old programmer about command lines
We're so back
Every dip in this graph is Rikten X waking up in the morning
(jk i appreciate all reviewers equally)
bbcbb70 Revert "add quetzal-keybinds (#11074)" - tylerwgrass
9f8e862 update configurable-slayer-task-overlay (#11654) - NathanVegetable
0f60840 update visibilityenhancer (#11702) - Endrit-alt
Summary
- reset the spec orb when passive special attack regeneration is actually observed from
SA_ENERGY - seed the last seen special energy value on startup to avoid false resyncs after enabling the plugin mid-cycle
- detect Lightbearer from the canonicalized ring item id so variations behave correctly
- add a targeted regression test for spec resync and Lightbearer variation handling
Testing
- `env JAVA_HOME=/tmp/jdk11/Contents/Home ./gradlew -g /tmp/gradle-home --no-daemon :clien...
Opened #20098 for this. This should fix the reported spec-meter desync by snapping the orb back to empty when passive special attack regeneration is actually observed from SA_ENERGY.
Opened #20098 for this as well. This likely fixes the older NMZ/Power Surge desync because it resyncs on observed SA_ENERGY increases instead of letting the meter free-run until full spec.
Opened #20098 for this as a possible fix. It might fix this report if the fast blue bar is the same underlying free-running spec-meter desync outside of Lightbearer scenarios.
Did the place that the version number for plugins get drawn from change? Its just showing the commit hash for my plugins now instead of the version in the gradle file.
Looked at another plugin and found that it's in the plugin properties now.
if you use build=standard it doesn't read your build.gradle at all, so it has to go in runelite-plugin.properties
How does the automated review work?
Ahhh okay that makes sense
Is there any way to read the version number from the runelite-plugin.properties file at runtime?
ExternalPluginManager.getDisplayData
it won't work for dev since its not external in the dev environment
What is the proper way to check if a trade window has opened/closed after WidgetID is deprecated
You can use net.runelite.api.gameval.InterfaceID instead
If you're in development mode and use the widget explorer, it will usually give you something inside InterfaceID you can use (e.g. Tradeconfirm.FRAME or something similar)
Ehhhh interesting thank you
If I have two plugins to update at the same time, I should keep the PR's separate, correct?
yes
gotcha, thanks for the help abex
Is there any solution to having accidentally turned off my code formatter at some point and now wanting to turn it back on which would cause a massive pointless PR diff
or is this just my life now
if you make a PR with only formatting changes, I can review it no problem. just ping me in here with the PR when you've made it
The RuneLite Wiki: https://github.com/runelite/runelite/wiki
You can build RuneLite locally using JDK 11. We recommend installing the JDK through IntelliJ and selecting vendor Eclipse Temurin (AdoptOpenJDK HotSpot) version 11. The codebase supports language features up to Java 11.
up to Java 11: Does this still hold to be true?
yes
Thanks
π₯
In LlemonDuck's Tombs of Amascut plugin, there is a "Swarmer Overlay" config option under the Kephri section, which overlays the swarm spawn number on top of each swarm. This is very useful to know when you can stop blowpiping swarms.
The final swarm that heals on 1st down is normally 14th, and on 2nd down is normally 18th. However, rarely one extra swarm can heal the boss (being 15th on 1st down, or 19th on 2nd down). My current understanding is that this is due to NPC index rollover, where the dynamic instance npc index goes over the ~65k max and wraps back around. This affects sub-tick processing order, which causes them to heal kephri. I would want to confirm this with some clips having NPC ID plugin enabled to prove that theory, but this is the current understanding. Same kind of thing happens with Vorkath flinch timing & zombie spawn index, for reference
Would it be allowed to add a special highlight if this situation is detected (based on Kephri's NPC index vs. swarm NPC index), to indicate to the player that they need to kill #15/#19?
One way of implementing it would be to highlight all swarms that need to be killed a special color (instead of the player just having to "know" and look at the numbers). That would be a separate enhancement. Then, the swarmer overlay could be made index-aware and detect the #15/#19 situation, and include those in the special color highlight should the situation arise
Not sure if that is within plugin rules
I'd think so, you should talk to llemon about it
@tulip wagon What do you think about that? My explanation ended up being two ideas:
- Highlight swarms that need to be killed a special color, instead of just numbers and having to understand what the numbers mean
- Detect the rare #15/#19 swarm index rollover situation, and include in above highlight
If you think it would be acceptable/within rules, but you're not interested in investing time in this, I could work on a PR for it as well
That's all fine, yeah
There's actually a similar thing already in I think tob qol which alerts when you are likely to get a NPC index rollover at maiden
To warm about scuffed crab spawn
@tulip wagon Ive accepted the changes you made in my PR 11535. Wanted to update you on that one
looks like it is indeed index rollover, like folks in the summit said
ill work on a PR to enhance the plugin sometime this week perhaps
I created the changes for this Kephri swarm discussion from earlier in this PR here, with some video examples of it working. Let me know if any changes are necessary: https://github.com/LlemonDuck/tombs-of-amascut/pull/139
@twin rover is there a way you could encode all this static data / mappings in json somehow, that way you wouldn't have to have the big diff
not even using gamevals for the npcs π
588d000 create hcim-guide-plugin - nandobrazil
bea7fc5 idle notifier: add dragonstone and onyx cutting - Adam-
hmm never done that before, ill look into it though for sure
claude/codex/w/e would probably be able to just scan it and figure out how to turn it into json w/ a loader
oh that might be easier, ill test that out later. thanks π
Ill also work on using gamevals, was learning by examples - not my fault all my examples didnt use it π
you should first ask codex / claude to evaluate some plugins that have been approved and the entire repo of the runelite master and plugintemplate that way it can understand the gamevals more properly. - usually newer models have enough context to take this in.
tbh adam, no, I had my model just copy the git locally and rip through it
we only added it i think yesterday'
Ehh ill prolly just do that part manually, just the json part with claude/codex
instead of wasting context, have it write a py script to write it locally maybe, just food for thought - take one file in, then can do many, rather than the agent doing it.
when I was looking for this, everyone said it wasn't possible
How long ago were you looking? It was added like 2 years ago
I was looking just this year, around april. I needed a setting-dependent toggle to hide/show relevant settings
that is a custom panel built into the config plugin, its not something you can make in a plugin
We don't support arbitrary setting panels, that is specific to notifications
yes, but this accomplishes what I'm looking to do
https://github.com/runelite/runelite/blob/master/runelite-client/src/main/java/net/runelite/client/plugins/woodcutting/WoodcuttingOverlay.java
Ah, I see how they did it now
?
wait, is this not the custom swing panel?
not even a little. and even if it was, we are they
how do you create the sub config swing panel?
?
then I guess I'm still confused. If this is baked into the config plugin, is it only the Notification class that has access to this type of sub menu?
yes
look at the config plugin
Hey Team, this is a proximity voice chat plugin that I have created for OSRS. I don't think I can capture/play user voice data with Runelite's limited API. https://github.com/runelite/plugin-hub/pull/11729/checks?check_run_id=73893732286 Any help or exceptions that can be made?
Hey, I'm interested in getting into contributing to some plugins or making my own, but I've always been curious about any involved 'risk' for my account? How does Jagex differentiate between me just developing, or if i'm running some unauthorized version of the client? If it somehow sees my account on a test client, how does it know i'm not abusing a test client to run unauthorized plugins in the name of 'development'? Or am i just worrying pointlessly? Apologies if this is already explained somewhere
just dont develop anything against the rules ^
is it normal to fail the plugin hub check using the json method? first time on it, passed the usual build.yml and it runs fine in IDE
b2adcd3 Update wiki data - RuneLite Wiki Scraper
Had to change like 4/5 things before it accepted the code.
you can't use reflection
Mine is a bit different.
Success π at least that AI conversion advice saved me lots of manual editing lol
using the javax sound stuff isn't strictly forbidden, that feedback is largely a mechanism to get people just playing clips to use AudioPlayer. for this usecase you'd have to use javax sound
so your submission is fine as-is and you'll just need to wait for someone to review it
@upper valve thanks brother. I donβt need donations or anything. Iβll sponsor the dedicated server myself. I just want to see how far itβll go. This will force me to start streaming OSRS again lol.
Trying to land another pentesting contract so itβll continue to fund this thing and Iβll donate more to RL.
Y'hello, whats the process to adopt an unsupported plugin?
https://github.com/Nerdpuff/max-skill-trim
I was in brief contact with the dev 2 months ago and they expressed some interest in handing it over but then did not respond further
@lean nova Pinging you in-case you happen to be available directly
That's the standard procedure
Is something like this good/clear enough?
https://github.com/Nerdpuff/max-skill-trim/issues/29
i think that's a fine step for 1.
I can't remember if this ever worked but the highlight for MOKI_AGIL_SHORTCUT_OP does not work. Only one pillar has the Jump-To op. Highlighting MOKI_AGIL_SHORTCUT_NORTH and MOKI_AGIL_SHORTCUT_SOUTH does work but it highlights both of them when you only need to click the one on the opposite side from you.
And MOKI_AGIL_SHORTCUT_MIDDLE highlights the actual pillar object
https://github.com/runelite/plugin-hub/pull/11678
any further changes needed for this or do I just have to wait through the whole PR review queue again
howdy - we'll ping you here or ask you in the linked PR when we need more information
it's not ID based through the UI, they have to right click hide each individual object like ground object hider
that is essentially a user-provided id list though
I think the thing with IDs is that people could target null objects
Like things you can't actually click or that don't have names
it's the same way previously plugins have implemented it, more or less
ground object hider is storing a list of IDs in the config
that plugin wouldnt be accepted if it was submitted today
it predates the no id based plugin rule
i guess we should remove it then
AI reviewer strikes again
i'm sure its better for everyone that it gets a proper review instead of your plugin getting bonked every time a rule breaking part of it is found
Are people who do config surgery disregarded in these cases? Because despite having to manually click the object to hide it, being ID/config based people could still hide null entities that way
what's an example of a null entity that's a no-no to hide?
iirc you werent allowed hiding something at bloat at some point
idr if still the case
a null npc that was bad to do something with(mark in this case) was vents at duke
i think it was the center chamber in bloat
because hide chains is still in tob qol
but yeah at any rate; im just saying even if you dont have a way to manually type IDs if it's saved in a way the user can config-edit it theres no real diff.
see : people highlighting all exhumes at xarpus without ever seeing all the variants
null entries need to be excluded in code
but wait a sec, if the center chamber is in the banned IDs list, how could they get around it in the config?
does not matter what people mark through config or in-game
im not saying they can, generalizing and was asking a question here regarding id based plugins. but yes you cant bypass those lists or null checks
yea, so looking at that code, I believe if they were to put the banned id in their config directly, when that function reads the config file it would get excluded
not fighting it, but imo this is such a weird line in the sand "you have to manually see the exhumed in each location before you're allowed to have them marked, but then it's OK"
never rly made sense to me
It's just a very literal interpretation of
Anything that automatically indicates where to stand, or not to stand. This applies to only automatic indicators, and not tiles which have been manually marked.
yea i see, I think jagex could use to refine that rule a little more π
I think the intention behind that is like, showing the AoE radius of some AoE attack, which requires understanding
Maybe miscomm here, it was just an example to show something requiring the user to manually mark it doesn't exactly prevent you from doing so via config-surgery which as seen people will do
My understanding is that you can do config surgery for both Object Marker and Custom Object Hider, so in that aspect, they are the same. As for banned things to hide, those can be disabled in the banned list (and manual config editting can't circumvent it)
The problem with ID plugins is new cases that were unnoticed, or only come up because of new content being released can result in doing something against Jagex rules. I assume thats why only old plugins that already had it were grandfathered in, so there's less plugins that can cause said issues to manage/keep track of.
Idk, assuming here but the specified reasoning on the rejected feature page is "moderation issues and outcomes that break Jagex's plugin rules"
Maybe I'm wrong, I'm not a contributor so more or less speculation on my part as to the reasoning.
@tulip wagon Ive added the deleted file back the pr was 11535
Type
Incorrect behavior
Operating System
Windows
OS Version
windows 10
Bug description
Kingdom of Miscellania overlay that tells you approval %
keeps disappearing after I hop worlds. reappears after a few more world hops
Screenshots or videos
No response
RuneLite version
runelite version: 1.12.24
launcher version: 2.6.8
Logs
[launcher.log](https://github.com/user-attachments...
@chrome crest@turbid turtle@novel stirrup fyi there is an api change next week where isTradeable() will mean tradeable between players and there is a new method isGeTradeable() for tradeable on the ge
Thanks for the notice! I'll make sure to update the plugin asap
Great thanks for the heads up!
oh the people that write loot filters will actually probably like that change
i didnt even know there were items that are tradeable that arent on the ge
Thanks for letting us know!
Burnt food among other things
the actual dig area is a 3x3 square and this is the center of that square
My prefix will benefit π
Map Clue steps have a 3x3 dig area in reality. This updates the highlighted tiles by the plugin to accurately point to the center of that dig area
@rancid marten or @amber rampart do you know why the existing highlight for MOKI_AGIL_SHORTCUT_OP doesn't highlight?
is it a multiloc?
Probably
When having the 'Press Enter to Chat' function enabled, the spacebar still works which is good, however it would be nice for the shift key to be active as well to e.g. drop items and tag stuff. Any opinions on adding that small feature?
It works just fine. You probably have another plugin that is stealing your shift key
If one of your keys isn't working in game, you've most likely bound it to a feature of a plugin. The Debug Helper hub plugin can show what keys are bound to what plugins.
(it's probably tasks tracker)
im trying to figure out what happened to my plugin which hooked script 2701 to do stuff with the prayer tooltip before it rendered.
it looks like it got abstracted into script 9444, but neither 2701 or 9444 fire when i mouse over a prayer to spawn the tooltip
what does fire is 8213 https://github.com/runelite/cs2-scripts/blob/master/scripts/[clientscript%2Cdeltooltip_timer].cs2 but using this, the full tool tip flashes before my plugin removes any of the text in the tooltip.
Any ideas whats going on here?
i just noticed 526 is also firing, using that seemed to work. i have no idea how that change happened or why it works though
what are you doing to change the text? prescriptfired?
onScriptPostFired
maybe pre fire would work for 8213
na same result for 8213 pre or post. ill just submit with 526 i guess
Quick question about an issue that was raised on one of my plugins (https://github.com/mhollink/bankstanding-plugin/issues/25)
When a player uses the !command, is it possible to show some data regarding the sender rather than the receiver?
we have a !kc service you can send an int through, but if you use it too heavily i might ask you to stop
Its for users to send and broadcast their bankstanding level... Not sure how often its used.
as long as its not automated in some way its probably fine
it looks like it has 393 installs so "almost never"
It currently prints the same message as !lvl <skill>. And is basically just that
Player types !bankstanding, thats the trigger
Is there an example which I can have a look at π³
the !kc code
Type
Incorrect behavior
Operating System
None
OS Version
No response
Bug description
'Bank', 'Examine', 'Ground Items' and 'Item Prices' plugins think the high alchemy value of the rune defender is 39k
But it can't be alchemised.
Screenshots or videos
No response
RuneLite version
RuneLite version: 1.12.24
Launcher version: 2.7.2
Logs
No response
Isn't this already a vanilla game setting
It is for spells but not prayers
?
the plugin adds more granularity to the options.
This is happening to me also, certain plugins will just not save the settings. For example, 117hd keeps being turned off when i launch. Others like combat attack style will revert to an old position. Tried reinstalling and using my old profile but that didn't fix it. At this point im thinking my profile is just corrupted and i'm gonna have to start a new one...
Y'hello, Im running into some issues trying to start my plugins
Looks like its not starting a graphical environment and then erroring out when theres no graphics
Any idea of where i should start looking?
My setup used to work a few months ago iirc
edit: tags for searchability [HeadlessException x11 wayland SplashScreen]
are you on a wayland wm/dm without xwayland installed/setup?
in a terminal:
echo "$WAYLAND_DISPLAY - $DISPLAY"
that looks correct
you can try restarting intellij through your terminal and see if that works, that might make intellij inherit your environment variables for now
if that doesn't work, in the intellij run configuration you can manually set the DISPLAY environment variable
gimme a secnond, turns out its been a while sine i ran apt upgrade xD
And i gotta make sure i patch copyfail while im at it
Are you building runelite on a server?
Nope, and i regurarly run rando scripts i find on the internet as root
But it still feels off to leave it unpatched
Maybe just don't run random scripts you found on the internet and you'll be fine until your distro has a patched kernel build ready to go 
(but also even without root, random internet scripts can cause a lot of trouble so maybe generally don't do that)
Fair fair
Same issue when running through flatpak from the terminal btw
Trying this now
That results in an immediate exit 1
can you try running intellij without flatpak?
It's installed through flatpak, i dont actually know how to run it without it π
i mean install without flatpak and run that instead. i suspect you're having flatpak issues - IDEs in flatpak are generally rough unless you know what you're doing
am i remembering correctly that idea community and the licenced one got merged into the same installation?
yes
Yo, we got a splashscreen, Thanks for the help everyone :D
I feel like im doing something wrong, trudging through the flatpak sandbox of the jagexlauncher is a liminal experience and a half.
How are you actually supposed to get the credentials.properties on linux
If itβs not in .runelite folder in ~/.runelite then itβs some where else
You may want to use bolt launcher, it makes a real file but also not in that folder location
It's more, how do i --configure runelite from a jagexlauncher install
I've found the exe (cursed thing to say when running linux) but its failing to start in --configure mode with a similar issue as above (DISPLAY settings)
https://github.com/runelite/runelite/wiki/RuneLite-Launcher-Configuration#runelite-launcher-configuration it should be done like this?
That only works with legacy accounts, no?
How do i get the jagexlauncher to run this instance of runelite?
The configure window works for Jagex accounts
How do i get the jagexlauncher to run this instance of runelite?
broadly speaking you don't
you get the credentials.properties file so a non-jagex launcher client can use a jagex account
Yeah i get that, but how do i configure the instance of runelite that jagexlauncher actually starts
find where the installed version of runelite lives, run its executable with --configure, then run the jagex launcher and launch runelite from there to write the credentials
Wait a second, do i just have an incredibly cursed install of the jagex launcher π€
its linux so probably
The βbestβ is just bolt launcher
bolt makes it a little less annoying
It's still not solved on my end. Have to manually prompt the locations by running back and forth. Runelite doesn't show any regions. Was able to solve the clue thanks to Wiki but otherwise was pretty challenging
It's still not solved on my end. Have to manually prompt the locations by running back and forth. Runelite doesn't show any regions. Was able to solve the clue thanks to Wiki but otherwise was pretty challenging
https://github.com/runelite/runelite/releases/tag/runelite-parent-1.12.24 -
that's because this was still not released yes, Runelite is still on 1.12.24 as of 3 weeks ago.
I've just noticed that my plugin "prestige-mode" is marked as broken and unavailable. (Commit reference: https://github.com/runelite/plugin-hub/commit/3664e60528036e743a9910dcadff05dd056d6742)
Running locally I don't see any issues, I'd like to try and understand why it's been disabled and what I can do to fix any issues that have been seen. Somebody has raised an issue in the github repo about RL crashing when checking a kebbit deadfall trap but I can't reproduce that locally. Any advice here would be very much appreciated!
Hi I'm super new to this, but I'd like to merge my plugin #11417.. Can anyone tell me what I need to do?
just gotta wait for someone to review it
ahh okay thank you!
we'll poke you in the PR itself or in discord if we have any questions
Thanks for sending that. Is there anywhere documented what needs to be run using clientThread.invoke - I did a quick search on google but couldn't find anything useful. I have a bunch of calls to client which seem to be okay (client.getWidget, client.getSpriteOverrides) but the call to client.queueChangedSkill seems to be causing the issue. Moving this inside a callback via clientThread.invoke appears to solve the problem.
not really. mostly its anything which mutates the client
I have just made a new PR https://github.com/runelite/plugin-hub/pull/11763 for a new plugin and I'm wondering if it is eligible for the new method of PR reviews based on the build=standard in the properties. I am confused why it says 'Requires maintainer review' in my PR, but not in a PR https://github.com/runelite/plugin-hub/pull/11754 for example that was added a bit before
Certain APIs can't be reviewed by the bot and still need maintainer review
What can I do to fix that?
for now, nothing
Wait for a reviewer
With my previous PR it was eligible for the automated review, seems like a bug in Github then? I will test to make a new PR to see if that fixes it. It makes a big difference to have it automated or having to wait a long time for a review
no, as stated you are using apis that need manual review
with what your plugin is doing you are not going to be able to get around the required maintainer review for now
also the automated review is broken right now so u really can't do anything to get it
I've found a similar functionality in key-remapping-plus, however that code also has the issue I ran into that both the spacebar and the rebind-button activate the click-here-to-continue. Whenever I rebind spacebar to something else, it completely stops working. I will try to fix that in key-remapping-plus
when are varps set by the client? is it after the gamestate is set to LOGGED_IN?
When the server sends them which is after LOGGED_IN
Thats what i thought, just confirmed looking at var inspector as well, ty
I believe this bug is related to #20010 though there is no NPE thrown.
The offending function seems to be isInKingdom() that contains a NPE check for client.getLocalPlayer()
I believe the infobox is never drawn as both calls to [processInfobox()](https://github.com/runelite/runelite/blob/master/runelite-client/src/main/java/net/run...
Regarding the above issue, when going about fixing this would the below be appropriate?
clientThread.invokeLater(() ->
{
final var localPlayer = client.getLocalPlayer();
if (localPlayer == null)
{
return false;
}
processInfobox();
return true;
});
Building RL on a new pc and running into this error: ```A problem occurred configuring project ':runelite-api'.
Could not resolve all dependencies for configuration ':runelite-api:classpath'.
Could not resolve project :runelite-gradle-plugin.
Required by:
project :runelite-api
Dependency requires at least JVM runtime version 11. This build uses a Java 8 JVM.```
I think I have it set to Java 11 in the project structure, not sure where it would be getting java 8 from
do you have a JAVA_HOME set perhaps
I don't think so, it's a relatively fresh windows install
do I need to set that to something reasonable? Or is that just a potential cause of the issue?
no, just it could be getting picked up somewhere and using that instead
I don't think I have JAVA_HOME set anywhere
Did you install Java11 from chocolately?
I think that app can set the Java home path but i donβt see it there either
I don't think so but I don't remember, I just followed the instructions in https://github.com/runelite/runelite/wiki/Building-with-IntelliJ-IDEA and in theory that should have downloaded the proper JDK?
In command prompt you type java -version you get 11? Or is it 8
Idk why it would let you pick 11 and then default to 8 in IntelliJ
well that's interesting
Thats sorta normal if its not set in path, idk if it needs to be for IntelliJ
You can set the path for jdk 11βs bin folder and see if it behaves but idk
Was just trying some stuff
do I need to install java separately from downloading the jdk through intellij?
Not in my experience no
how can I tell if I have java installed? Should I have a C:\Program Files\Java folder or is that not necessary? I can see a C:\Users\bigdi\.jdks\temurin-11.0.31 folder
IntelliJ puts it in that second spot I think
And no it doesnβt matter where it is, you install it anywhere and then your programs are told where to look for it. IntelliJ should be doing that automagically
I don't see that .jdks folder in my PATH, should I add that somewhere?
You can try that yeah
Donβt see why not
I think you want the bin folder
Java/version11/bin
like so?
If you reload cmd and type Java -version it should show 11 then
and you may want the system env, not user?
alright cool, looks like that's a bit happier
lol
@shy pine see https://github.com/runelite/plugin-hub/pull/11776#issuecomment-4379378151 for your next release if you care
Still running into some errors, I'm following the "how to build" guide, I get to the section that says "Select Run RuneLite.main() and wait a while for the client to build and start." but I don't see those options, where should I be looking?
reload gradle project is next thing i would try
in build execution deployment -> build tools -> gradle make sure it looks like
The gradle jvm is a separate setting, (from memory) something like: settings -> build tools -> gradle -> Adams screenshot above
oh yeah we're cooking now
well that doesn't seem great, but at least it's a new thing
Your gradle still looks odd in that picture
gradle settings look okay?
I created the PR to takeover the scrollboxcounter plugin, changing the git repo to mine but leaving the commit hash the same. I noticed that the build is failing, did I do something wrong here? https://github.com/runelite/plugin-hub/pull/11777
The right side is blank
In the picture of the build
I see that, how do I get it to not be that
You've followed the instructions correctly - for older plugins you'll need to actually change the commit hash at the same time to update the runelite-plugin.properties file to include the build= option
Hmm, is that true? Seems to look ok here. Thanks for the review btw
oh
maybe i'm dank
I do see some folks putting their build version in their runelite-plugin.properties file but this seems very inconsistent
oh yeah nvm you're using build=gradle, not build=standard, you're all good
i think you should be able to use build=standard going forward, but it would require you to rethink the version.txt feature. i think abex was looking to expose the version that's included in runelite-plugin.properties file
I was just about to ask specifically about that hahaha
I already have a bug fix on my fork, do I need to force push to remove that to include the build= option? Or can I ship both at once
I would strongly prefer if you only push the build= option commit
What do I specifically need after the build= btw?
You can make a new branch off the 98c1ec30e075be7b37a01be53081908746dc1db3 commit and include the build=standard change and push the commit hash from that branch, that way you don't have to do a force push
Yeah it's important for me to track the version for compatibility checks on older saved files in Creator's Kit. Would be very nice if it weren't this... strange to accomplish that
as long as you're not doing anything weird with your build.gradle file, like including dependencies that aren't in the template file or adding additional build steps.you can use build=standard
so like this: https://github.com/RonnyTommy548/scrollboxcounter/commits/build-fix/ and also change commit hash in my plugin hub PR to 9b4071e39dc40d6d120fbdb58be3287bf1154026, despite what the takeover instructions say, as a special exception due to old plugin?
i would've personally included a String VERSION = "1.2.3"; in your main plugin java file and add a workflow/document to ensure you update it in tandem with plugin updates
yep!
5a00892 API docs for 1.12.25 - RuneLite updater
31e7343 Release 1.12.25 - RuneLite updater
Ah yeah that's something I wanted to avoid, but was probably the best route from the beginning. Especially so now with RuneliteGPT
i think just keep build=gradle as-is for now until you can read the version from one place
i know others would like to be able to read it from their plugins too
Sounds good
Type
Other
Operating System
Windows
OS Version
I use Jagex launcher to launch Runelite
Bug description
No matter if i turn it off or back on the weight of items doesn't show nor does the stats of gear shows even when check marked its weird, because when i launch regular runelite it works for my other accounts but not when i launch runelite through jagex launcher,I'm not sure if others are experiencing it but i am not sure why its not working when check marked for myself...
67c0de8 API docs for 1.12.26 - RuneLite updater
2cefa3b Release 1.12.26 - RuneLite updater
related to dayohs takeover request, any opinions on this to avoid one more misdirection from happening? https://github.com/runelite/wiki/issues/173
https://github.com/runelite/wiki/blob/ae4327009d1853de4413e16d9aaea296c4d6733d/Plugin takeover policy.md Current step 2 instructs not to change the commit hash, but it could be move beneficial ...
I think at one point there was a problem where we couldn't reliably diff plugins between repos if the repo + also plugin changed at the same time, is that still a thing?
It's not, no. I think it's still clean to keep the transfer PR separate from any unimportant changes
The support URL in the runelite-plugin.properties is not used iirc, we always link to the github repo's issues page.
I'm not sure updating the author field is really correct - the original plugin owner is still the author/an author.
support url pointed to the old repo until i changed that file when i took over protect-item-notify π€
As in, in the plugin sidebar, you right-clicked & clicked "Support"? or on the plugin hub website https://runelite.net/plugin-hub/show/quest-helper
website, cant remember if i tried from the sidebar
probably because in take overs people didn't change it π
hmm yeah, weird https://github.com/runelite/runelite.net/blob/master/src/routes/plugin-hub-show.js#L79
RuneLite website. Contribute to runelite/runelite.net development by creating an account on GitHub.
e11d20e update daily-agility - DrieVeertien
0543eb9 create rune-kite-streak - anthonyal2k
Type
Incorrect behavior
Operating System
Windows
OS Version
Windows 11 build 22621
Bug description
The anchor points above the chat are broken. The left one is misaligned when the chat is closed:
while having the chat open, does align the left achor point most left:
It does not matter if the chat is located at the bottom left, when the chatbox interface is moved
, the same behaviour applies.
It seems to be present in both the resizable layouts. Disabling the stre...
eafc7f0 update clue-teleport-helper - kitten-lissy
2ec4e40 create object-marker-plus - Shackin
4f79b3a create zmi-helper - NathanVegetable
back from lunch and trying to get dev to build again π how do I get the gradle thing to actually do its thing?
[ https://github.com/damencs ] [ @tidal stag 241037329408458752 ] damen#0
@tidal stag seen a few people report some new issue with 238 and tob-qol, can you look into it now?
close the project, delete the .idea folder in it, re-open, it should auto-detect gradle and load correctly
gave that a spin but now it's running into a dependency issue, I think?
I can take a look. Any general direction of issue? Doesn't load at ALL or? I haven't seen any reports personally.
bloat chains not being removed
I think it might be something with removeGameObject() not working
Thanks Coop.
I think we see the issue and it is just removeGameObject() isnt working
I'm talking to someone that is already taking a peek and it is successfully removing chains, but there are 4 that aren't. I'm watching a fresh TOB run to see comparison of it.
it would only happen with gpu on i think
The method is working WITHOUT gpu.
yeah
This was written an eternity ago, so I am happy to realign to necessary approaches.. if you guys are refactoring things intentionally around this area.
its definitely a client bug i just didnt know what was wrong with it
Let me know if you need any additional information or want testing again on a new version.
Nylo Pillars are NPCs, so that should be treated differently.. so it spans beyond Game Objects - fyi.
Sorry.. refreshing myself a splash. The pillars are "npcs" but also game objects. The pillars are hidden as a game object.. with the same method as the bloat chains.
are you sure they aren't game objects with an invisible npc for the healthbar?
to be clear you are getting the GameObject which has .getRenderable() instanceof NPC?
and removing that?
he is removing the loc TOB_NYLOCAS_SUPPORT_PRISTINE
so its the same thing as the bloat chains
yeah, ok
Correct.
It is just odd that SOME of them are removed.. but not all.
With GPU off.. NONE of them show.
gpu off uses a different renderer
GPU is on for that screenshot above.
Just trying to give you information that maybe is useful. I'm not even sure what has been changed between the versions.
@tidal stag can you test latest.integration in like 10 minutes?
Pulling and building. Will keep you posted.
make sure it says patch 25397377789.125
It's working now.
For bloat. Testing Nylo as well after bloat.
Nylo is good as well.
ty
Anytime. Thanks for the ping.
If i have a minor code change to a plugin, and documentation changes, is it preferred to update the plugin in two steps (one with just code, one with documentation) to make it easier to review, or keep it as one review to not spam?
if it's code change + documentation change, feel free to leave it as a single PR
doing some more experimenting with rlobjects tonight, drawing them isn't expensive right? lets say just some random model from cache, cloning it and maybe modifying the transparency of specific vertices
cant test yet but just curious if they behave whole lot differently compared to java2d ^^
They're rendered in the same way that other in-game objects are so, they're not more costly than any other object in the scene
The API around them (including animations) are pretty thin wrappers around the corresponding internals
I'm attempting to use this method to play positional sound effects: https://static.runelite.net/runelite-api/apidocs/net/runelite/api/Client.html#playSoundEffect(int,int,int,int,int)
But I've had no success, so far. I've even tried manually passing in coords from the debug menu to originate from my players location. I'm hoping I'm just passing in incorrect values, and that's why it isn't working.
The docs say @param x the ground coordinate on the x axis I'm not sure which coordinate system it means, I must be missing something
I tried Local, World, and Base coodinates, but there's still no audio. If I use playSoundEffect(id, volume) it works, it's just no longer positional audio
If anyone has any ideas, please let me know
they are scene coords I think
That doesn't seem to work either
I'm using public void onAreaSoundEffectPlayed(AreaSoundEffectPlayed event) and logging the event's scene x and y, and they come out as 54, 49. Trying to replay the sound with the coords does not work
I'm assuming the api isn't broken, but I feel like I've exhausted my options
it works for me
What are you passing in for range and delay?
1, 0
would someone be able to check this hot fix for me please?
https://github.com/runelite/plugin-hub/pull/11780
Maybe it has something to do with the event being consumed before I try playing the sound? This is strange
you aren't consuming your own sound are you?
ty Abex β€οΈ
Maybe I am if this doesn't behave the same way normal sound events work? I have it simplified down to this for debugging:
@Subscribe
public void onAreaSoundEffectPlayed(AreaSoundEffectPlayed event) {
event.consume();
client.playSoundEffect(2562, 50, 65, 1, 0);
}
But if I change that line to client.playSoundEffect(2562, 127); I can hear the sound just fine
Maybe if it's like queuing the sound effect when you use positional audio, it actually cancels the sound?
you have area sounds unmuted right?
Good idea I'll try that out after dinner. Thanks
and you are going to be consuming your own soundPlayed event
Right it'll trigger this function again, interesting
Curious how that should be handled
probably just set a flag in your plugin to ignore the next event
its guaranteed to be synchronous
Having the same problem.
49fba7f API docs for 1.12.26.1 - RuneLite updater
92a8cbe Release 1.12.26.1 - RuneLite updater
hmm bridges
Oh still on the old profile, works on ground level but not when you step up to the stone floor @ GE - still thinks you're on ground. Same with the new profile.
At some point in the past the distance which you could see NPCs was decreased. Although I've noticed at tempoross you can see the jumping fish animation although the fishing plugin doesn't highlight it and onNpcSpawned events don't fire until you get closer. Is there a restriction in npc events? Or does the client know about animations from a further distance compared to NPCs?
Same here, happened to me just now. Dropping to ~30 fps. If I turn the GPU plugin off, it goes down to ~10. Not sure if it is the same problem.
i dont think the distance you can see npcs has ever changed in the history of osrs
I don't know the exact date but NPC Idle Timer used to show on the barb fishing spots when you banked at CoX bank. But now when you run that same distance the NPCs unload
weird
the tile height change in 1.12.26.1 seems to have broken quite a few things
i blame @low hatch
does it affect ground items?
Yeah
the text clips through the item model
I am watching a stream people complaining to me about this
Might be unrelated then or a hub plugin I canβt tell
The player indicators though is quite annoying though
I donβt remember if that is core or not
it does seem like a regression
looks pretty ok at least not on a bridge
yeah looks good in sote room
except on the stairs before the room
Elevated surfaces i guess
id believe that, i dont believe ground items are broken
It just seems odd to me you can see the leaping fish which is TEMPOROSS_HARPOONFISH_FISHINGSPOT_SPECIAL but it isn't tracked as an NPC. But maybe that is expected
devtools overlay might be culling it
@rancid marten if u push to maven i can test
But the fishing plugin doesn't highlight it either. Neither does the Tempoross plugin notify for it
maybe its a spotanim or loc or something
fixes it for NPCs too ^ (previously nylos walking down the lane would have a similar bug)
@syunko5 I have a similar issue.
I was able to work around the issue along with the other UI positioning issues by setting "Enable Custom Window Chrome" to true in RuneLite's settings.
<img width="255" height="39" alt="Image" src="https://github.com/user-attachments/assets/571c8cb1-4edc-403c-8cc0-09732572a4ba" />
Hey hey I didn't ask to make the changes π
A small update to the Item Prices plugin to allow users to disable the overlay for specific items. This is useful for items like potions or food, for which you typically only care about what they'll do for your stats, not what they're worth.
I referenced the Ground Items plugin for the implementation here.
Functional verification:
-
prices still shown when an item is not disabled:
-
price overlay hidden when an item is disabled:
-
price overlay hidden when an item is disabled via...
782dd40 Update jav_config.ws - RuneLite updater
Extend Anti Drag plugin to the Rune Pouch interface. Useful in certain high end PVM scenarios where fast trades are required.
Dont want to be a pest but any update on the camera issues? Its rather annoying π
can you be more specific?
The top down one and the jitter when doing certain leagues animations, tbf im still at work so im unaware if its been addressed
i think that was fixed yesterday, have you checked since yesterday?
Barb gathering calcified rocks looks fine for me
I have not, sorry if it was a stupid question π
I don't know if this is a Jagex client or runelite issue, but Actor#getInteracting() appears to never clear after today's update
yeah jagex broke sometihng
Jagex issue
Camera is still forced to like an up-down angle at bridges
even with extended pitch limit off
it's different to what it used to be
i dont really believe that / if you want me to do anything other than tell you to go away, provide specific evidence
Maybe it's the pitch limit being off
It just felt weird the entire day but I guess it might have been the pitch limit extension being off
all seems normal now on 1.12.26.1, regarding both the extreme pitch angles and the camera snapping down on certain animations
Type
Incorrect behavior
Operating System
Linux
OS Version
CachyOS
Bug description
bug occurs regardless if I'm on GPU/117 HD/or default renderer
Bug also occurs on other plugins that highlight npcs, like player outline.
Bug also occurs on a completely fresh profile and fresh install.
Screenshots or videos
RuneLite version
Runelite version: 1.12.26.1
Launcher version: 2.7.7
Logs
No response
3d60a79 Revert "api: fix computing getFootprintTileHeight" - Adam-
Same issue but even worse inside the hunter's guild. All npc highlights are not in sync with npc/creature; instead they are below. Impacts tag highlights, and hunter rumours plugin so I think it's a general client issue.
<img width="403" height="405" alt="Image" src="https://github.com/user-attachments/assets/96f5b662-8570-4eb2-9249-32c520213f2a" />
I think I just ran across this as well, though it could be a similar but separate issue
<img width="119" height="199" alt="Image" src="https://github.com/user-attachments/assets/665fe4e7-7499-462d-9900-4083190c181c" />
I noticed it on docks specifically, but it could be anywhere that the object is on a "floor" that's offset from the terrain?
RL version: 1.12.26.1
OS: windows 10
Seen on region locker GPU plugin.
I think I've been running into this issue, though it was a different aspect that I initially noticed. (NPCs remaining highlighted too long after being interacted with/attacked)
RL version 1.12.26.1
https://github.com/user-attachments/assets/11dcfa3c-ad17-4b95-a6a0-46df7d2497bd
Fix for #20113
Original LocalPoint-based height sampling (which did bilinear interpolation and handled bridge/scene coordinates) replaced with direct WorldView.getTileHeight() calls, bypassing interpolation and bridge-aware adjustments and producing incorrectly low, misaligned highlights.
This reverts the recent getFootprintTileHeight changes in Perspective.java and restores the original LocalPoint-based height sampling.
This brings actor highlight height calculations back in sync w...
You're correct - nice work, my bad!
I'm working on this.
@waynesan41, are you still working on this? If not, I would love to have a go at itπ
a6bec8b API docs for 1.12.26.2 - RuneLite updater
d76033b Release 1.12.26.2 - RuneLite updater
i am also fixing the facing moonwalking thing with this i think but I am not very super confident in it
so I guess lmk
4f09415 update gauntlet-chest-popup - ldavid432
5979dfd create inventory-highlighter - Emtec-byte
bba1beb create logout-fix-resizable-modern - psikoo
Type
Incorrect behavior
Operating System
Windows
OS Version
windows 11
Bug description
title, after weekly reset this bug as appeared. also happens in safe mode
Screenshots or videos
RuneLite version
1.12.26.2
2.7.6
Logs
b0e6c29 update volume-control - avahe-kellenberger
eaffce5 update bone-mining-notifier - staticvoid07
98bd4f4 update barbarian-fishing-timers - staticvoid07
happened to me as well today
Could you please make an updated graph?
bit hard to tell, but PRs are going up again?
faster review, so maybe people submit more PRs?
yeah a lot of people cant think more than a few minutes ahead and will submit multiple a day
also it was offline for like 3 days earlier this week due to an issue
induced demand strikes again
I opened a relatively small PR on lemon duck's ToA plugin 1 week ago, which we briefly discussed & agreed on in discord. I've been looking forward to a review on that. However, I'm starting to realize that this is pitted against all of lemon duck's other work, like reviewing runelite plugin hub PRs, and other plugins that lemon owns, other longer-standing or more important issues/PRs within those plugins. I was thinking about pinging lemon about my PR but I don't want to come off as a nuisance, seeing how my request is one of many.
What do people normally do in this kind of development scenario? This one is a little different than my situation with scrollboxcounter plugin, where that guy is MIA and I had to initiate the takeover process to get my fix through. Lemon duck just seems to have a lot of work going on. Is there a way to see if there are other collaborators on the ToA plugin I could ask? If not I guess I just have to wait patiently, maybe remind after multiple weeks incase it's forgotten. Just thinking out loud here
yes ping llemon, she is around
@tulip wagon Do you know when I might expect a review on PR #139 for Tombs of Amascut plugin? It's that idea I brought up last week for only labelling healing swarms (w/ index rollover logic), fairly small/straightforward diff ~50 lines. All good if you're not able to get around to it for a while, just looking to understand/make sure it wasn't forgotten
51be286 update barbarian-fishing-timers - staticvoid07
Same issue using the "House Thieving Varlamore" plugin.
npc's stay marked as distracted after the distraction is over.
<img width="682" height="502" alt="Image" src="https://github.com/user-attachments/assets/d074cb09-0e89-4a4a-beba-9678939753a0" />
OS
Tahoe 26.3.1
Mac M5
RuneLite version
1.12.36.2
2.7.6
Oldschool revision 238
Behavior is replicable across sessions and restarts.
It's a jagex bug since the 6 May 2026 update. You'll just have to wait for Jagex to fix NPC interactions.
Is this part actually a jagex bug or just getInteracting() not accounting for when it's reset
people keep saying its a jagex bug but the character isnt visibly facing the thing
so i dont understand that really
With fishing spots the character turns to face the fishing spot after it has moved
It's a jagex bug since the 6 May 2026 update. You'll just have to wait for Jagex to fix NPC interactions.
Is there any other source for this?
I trust you I just want to watch it closely.
a rapid way to see this is at minnows
the fishing thing seems like it would be unrelated
if you arent facing it then you just arent facing it
like theres no way that this explanation makes sense
i thought it was because you're interacting with it without moving towards it, like you can interact with players by using a spell on them when at the ge
#support message this guy had an opinion but idk who he is
Well then there's a runelite API bug with interacting but I'm not able to determine if that's the case
who the fuck is this
[[Isles]]
Another example, after unnoting bones on Virilis and then blessing them on the exposed altar it still shows you're interacting with the NPC
do you know what it does when it resets
uhm yeah sec
bb.ag is set to lf.ak, but the other fields are not cleared which is why it's broken atm
oh i see
I am also noticing this bug in happening different places
lol yeah idk who that is
Given artefact lures were broken I just assumed it was all NPC-player interactions and not a runelite api issue
there is a mix of rl api broken and also the server is broken atm
which makes everything confusing
and also the clientis broken
basically, everything is broken in different ways
actually to clarify, it's set to lf.ak when facing an entity so really just need to check if that field is equal to it in the api I guess
perfect then
i dont understand why this is so complicated
yeah these facing changes are def something
the server also faces objects incorrectly since it says the size of them are always 1x1 for some reason (though that was broken in 237 server sided too)
so you just always face SW tile
was it broken in 237 or did you just notice it there
it was broken in 237 too, but they calced the angle server sided
pretty sure it broke in 237
do you have a good example
Would that be the odd pathing problem people noticed?
i assume it broke in 237 as prep work for these facing changes, but somehow they ended up discarding the size of objects along the way
but who knows lol
we never got a repro for the pathing thing did we
idk im being gaslit pretty hard i think the last day or two
yday someone in #support tried to convince me you are supposed to be able to move your camera below terrain
i looked into it a bit and i really dont see anything wrong
below... idk about that lol
even if you could, why would someone want to do that
would this be a good time to tell you another bug (cache related) to pass on
sure
for some reason there's some sounds being repacked every update since 237
with new ids
while keeping the old
so its duplicating the sounds?
yup
do you know which ids
the latest ones are 11505 11506
i have a theory that it's because they both end the name with a space
that sounds stupid enough to be true
they have these dbrows that i assume are generated from disk somehow at one point
and this entry (and another one) keep changing ids
and notice the space at the end.. π
yeah
ive been discovered
do you have more info on this? like what the issue actually is
no i think its not real
essentially every day we have people come into #support and make fucking insane claims
and occasionally they do that on the same day as engine updates
and it makes my life hard
because there is a weird one if you do an interaction then spam click to do another on some other thing
but idk if it's referring to that
idk
i saw a thread on reddit about it let me see if i can find it
This was what I was thinking of https://reddit.com/r/2007scape/comments/1t5f1ut/new_vorkath_pathing_issues/
but some of the other people were insisting its been happening for over a week which means it not related to 238
but ive not been able to get an actual info about it
oh maybe not, that vid is from yesterday
yup, is fixed
i liked it more when everything was just ints
yeah i feel like their refactors are moving things to new objects and using enums a lot more
https://www.reddit.com/r/2007scape/comments/1t6egf6/jagex_priorities/ even this says pathing is broken since leagues ?? what does that mean
Iβve also seen this about follow pathing. But suhdude in the replies says itβs been happening since Brutus or Sailing so idk https://x.com/eliopdagod/status/2050075510789882216
it happpened a lot in dmm
i noticed this exact pathing in dmm
thojught it was a pvp thing lol
Happening to me as well as of today.
thanks for the reminder, i'll try to look at it today
Summary
- Track bank, shared bank, and seed vault interface lifecycle before honoring the bank search shortcut
- Ignore the shortcut while bank search input is already active so the bound key can be typed normally
- Consume only the paired typed event from opening search, preventing plain-letter shortcuts from seeding the search text
- Add BankPlugin tests for open, active-search, closed-bank, and opening-character cases
Fixes #14342
Addresses the bank-search portion of #13608.
Testin...
14e002b update prayer-time-infobox (#11798) - arbeerby-pixel
Thank you
40f3c6b update drop-rate-properties - Pluckss
It's a jagex bug since the 6 May 2026 update. You'll just have to wait for Jagex to fix NPC interactions.
How is it a jagex bug when you can disable the interact highlight plugin and it goes away?
what was the conclusion on this u and polar started talking in symbols so i stopped reading
It's actually a runelite internal bug that Adam fixed
litty
How is it a jagex bug when you can disable the interact highlight plugin and it goes away?
It was assumed because of a different Jagex bug, that the interact state was not being correctly reset, but it was confirmed to actually be a RuneLite bug.
@RAKEBangalang it appears fixed for me now.
@RAKEBangalang it appears fixed for me now.
Are you using npc highlight or a different plugin? Plus youβre on Mac so itβs an entirely different issue
@RAKEBangalang it appears fixed for me now.
Are you using npc highlight or a different plugin? Plus youβre on Mac so itβs an entirely different issue
I'm using npc highlight. I don't think there is a seperate mac runelite client it's built from the same source.
To clarify though, you're still seeing the issue on your machine?
that guy really said do not speak to me apple scum
dont think there's been another release with the fix yet either lol
@rancid marten did you add the same check to Actor#isInteracting too?
idk it's unused in core, you added it π
idk
Leave it to the plugins to implement haha https://github.com/m0bilebtw/c-engineer-completed/blob/1afd76d4a06cb9b904fe67436dd6953ca8b11371/src/main/java/com/github/m0bilebtw/player/CEngineerPlayer.java#L140-L148
well this returns true if you are interacting with something you cant see
which idk if that happens
maybe
clicking an object and npc in a very short time span causes interact highlight to get confused and highlight both
3daebee Update wiki data - RuneLite Wiki Scraper
Type
Incorrect behavior
Operating System
Windows
OS Version
Windows 10 IoT Enterprise LTSC
Bug description
While not strictly identical, I think the game messages you get in leagues telling you how many of an item you have banked as it sends it to the bank for you should be collapsed by this option.
Maybe lines could be considered duplicate if the only difference between them is a number?
Screenshots or videos
RuneLite version
RuneLite version: 1.12.26.2...
https://github.com/runelite/runelite/pull/19978 Anyone able to take a quick look at this sometime?
@RAKEBangalang it appears fixed for me now.
Are you using npc highlight or a different plugin? Plus youβre on Mac so itβs an entirely different issue
I'm using npc highlight. I don't think there is a seperate mac runelite client it's built from the same source.
To clarify though, you're still seeing the issue on your machine?
yes, still persistent.
is it allowed to store a click that failed due to things such as animations/cooldowns and display that to the player so they know the click never registered?
sounds fine to show feedback like that
you just cant store it and fire it off again when it would have worked
also question about prayer flicking helper
if I could detect the pattern of attack per enemy and display that as a indicator when to click the prayer would that be allowed?
No
Jagex's third party client guidelines: https://secure.runescape.com/m=news/third-party-client-guidelines?oldschool=1
Additionally, these features have been rejected or removed from RuneLite: https://github.com/runelite/runelite/wiki/Rejected-or-Rolled-Back-Features
Indicates what prayers to use in what order, for example in the Cerberus boss fight
it seems to be boss focused, but im guessing its just the example they used and it applies to all enemies as well?
Indicates what prayers to use in what order
Does not say its a boss restriction, it says combat
cool, well thats good
so even if it didnt state what prayer to use and only gave you a window of time indicator it would still be disallowed?
Cerb was just a good example since most people know about that effect
yes
I'd say that's still bad
gotcha
- Next attack prediction (timing or attack style)
Yeah looks like this covers it
29c64c5 create skillingloottracker - daviddavies9897-wq
e60cd47 create placeholder-prices - Maximuis94
entity hider can hide 2d
Normally don't call out plugin PRs, but we had an aggressive default keybind that I've removed (shift + space), among other very small changes.
Sorry about that ...
https://github.com/runelite/plugin-hub/pull/11833
this is gonna reset it for everyone that changed it themselves though
i merged 
but that still would be better than how it currently is
I think thats a good thing in this case
that was my thought too
yeah that was my opinion as well
yeah I was thinking the same thing (i wasnt)
its when i went to type S I T that i realized how bad it really was
the ol lesser of two evils play
the people that changed the keybind know its a keybind so they will find it again 
Could've written a migration if the config wasn't shift+space but oh well
am I supposed to have a role once my plugin is approved and on runelite?
It requires a certain number of commits to the plugin-hub repository for the dark purple name colour
Github API only returns the top 500 contributors by commit count
I think the barrier is... 7 commits at the moment?
oh thats odd lol
They can't help the Github API's limitations in that regard
would kinda be cool to have roles based on how many active users their plugins had
lol
1475b53 update shortest-path to v1.19.2 (#11830) - FIrgolitsch
If you're AFK fishing and the inventory fills up, the idle notification doesn't fire while the fishing spot is still there. It only fires if the spot moves.
checkInteractionIdle only triggers when Player.getInteracting() is null. The fishing spot NPC is what you're interacting with while you fish. When the inventory fills you stop fishing, but the game client still has the fishing spot as your target, so lastInteracting gets refreshed every tick and the timer never elapses.
Adds a che...
The getInteracting API will be fixed in the next runelite update. It was a regression from Wednesday's client/engine update.
Didn't realize this was already in the works. Closing this since the API fix will cover it.
I discovered that the issue that is causing the Interact Highlight bug is the Menu Entry Swapper plugin, specifically the option "Bank" from "NPC swaps". When turned off, the highlight doesn't stick to the NPC anymore, however, a full logout is required to fix it, as it appears to persist even when the plugin is turned off in-game.
0e8687c update trueclueareas - Daniel-Tudose
889b223 update shortest-path to v1.19.3 (#11843) - FIrgolitsch
eaba489 API docs for 1.12.26.3 - RuneLite updater
7cd6829 Release 1.12.26.3 - RuneLite updater
Why this is such a nice QoL for bank tag tabs for players that have a high count of bank tag tabs:
With my play style, I frequently have to update the auto layout of my bank tag tabs.
The tag tab scroll bar has the "Auto layout" options. However the "View tag tabs" UI does not have this.
At a high count of tag tabs it becomes more efficient to navigate between tabs using the "View tag tabs" UI rather than scrolling the tag tab bar. But this means you still have to scroll to the tab u...
a2f837b update potion-storage-customizer - ASong5
c6b3cf0 update boss-respawn-notifier - ASong5
Lowkey wish the feature was in core π
The problem is getting someone to test all of the exact areas as opposed to just using the worldpoint from cache
yes putting info that the plugin is "not sure about" is not a valid option for core plugins
Obviously π I didn't mean that plugin in its state. I meant the boundaries feature generally π