What's a good channel scanner?

A problem has come up with nuisance juveniles using a device that makes it appear that a nearby avatar is saying embarrassing or obscene things - and no, the text is not in tell-tale green, but in white like any other chat message. To operate the thing, they have to give it a command in a concealed channel. Scan that channel and you see the command, see who is issuing the command, and therefore know who to AR and otherwise deal with. But there are so many channels, so little time.

What are the good channel scanners out there? Is there one that really stands out? Inquiring minds need to know.

Hm. I don't know how that can be done by script, and if it can be done by packet injection, I wouldn't expect it to be controlled via chat... not that I'm calling you a liar, I'm just interested to hear how it might be done.

I don't think objects can do white text... Are you _sure_ the person that it appears to come from isn't playing games with you or doing a gesture?

Positive. I have seen this phenomenon with chat-spamming griefer objects in public sandboxes, too, so it's around. But now it has turned up in clubs, where it is much more disruptive. And yes, the text is white.

I'd be awfully surprised if it could. I used to have code that could do that via SLProxy, but LL closed that little hole quite some time ago (maybe a year ago?).

If it's in white text, then I would definitely suggest that the OP submit an Abuse Report, make LL aware of the situation. Perhaps they can go over server-side data and find out who's doing the packet injection (or whatever they are doing). This is a VERY SERIOUS exploit if it's true. Imagine being able to make avatars give their vendors commands without their knowledge, etc.

Don't just look for a channel scanner, and don't wait! If it's done via packet injection, they only have to be in the same sim, they do not have to be in scanner range at all. This needs to be brought to LL's attention IMMEDIATELY. I'm sure they would rather you err on the side of caution in such a case.

There are 4 billion channels to scan. Good luck with that!

I would suggest filing an Abuse Report for each instance, make sure LL knows that is happening. I know I probably sound alarmist, but if it's truly mimicking people, it's a cause for concern. I doubt that it truly is, but LL is in the best position to make that determination.

Most probable explanation: An object (worn in a no-rezz environment) that can be given the name of the target avatar. Dunno if it could trigger a victim's vendor devices or not. The vast majority of "secret" channels seem to fall in the range of channel numbers 1 - 999 for the obvious reason that nobody wants to type in a lot of digits when giving a command. Will probably AR at next opportunity, but meanwhile...

...Anybody have any suggestions for a good channel scanner?

I don't have any suggestions, never having used one, but I know there are quite a few listed on SL Exchange. I imagine that they all work pretty much the same way, so that's as good a way to select among them as any, I suppose.

I've never seen (and can't quite believe in) an object capable of causing your viewer to display white text. I agree that it's probably innocuous, but I still recommend reporting it

Best of luck to you, I hope you succeed in getting rid of the problem.

I think Mystitools has a channel scanner or something close. I might be wrong but I think it does. Good luck!

I will ask about this on the scripters' and sldev lists.

How do you know it is being chat controlled?

On my viewer (OnRez) the colors of chat aren't the same.

What should be yellow shows up as white.

And what should be green shows up as ever-so-slightly gray.

OP - Are you using OnRez by any chance? Because if you are, it is entirely possible that the text is green to everyone but you and those who are also using OnRez.

Don't have OnRez, and the text is white to all present.

And Mystitool is wonderful in many ways, but turns out it can only monitor 32 channels at a time. : (

So much for trying to find a silver lining. I hope you can find a solution so that LL can stop the exploit.

If this stuff is happening in open chat, and you are certain it's coming from an object, then you could use a simple script to tell you the key of the speaker, and the name and key of the speaker's owner. No need for a channel scanner or listening to 4 billion channels.

The only thing is while you have this script active it will 'spam' you with the details for everything that anyone says. But if you put it in an attachment then you could just wear it when the problem was happening.

default {
    on_rez(integer n) {
        llResetScript();
    }
    state_entry() {
        llListen(0, "", NULL_KEY, "");
    }
    listen(integer c, string n, key i, string m) {
        key ownedByKey = llGetOwnerKey(i);
        if((i == llGetOwner()) || (ownedByKey==llGetOwner())) return;
        string ownedByName = llKey2Name(ownedByKey);
        llOwnerSay("Speaker: " + n + " - " + (string)i + " owned by " + ownedByName + " - " + (string)ownedByKey);
    }
}

Every time anyone or thing (other than you or your objects) speaks, you will get in private chat a message with the name of the speaker, the key of the speaker, and the name and key of the speaker's owner, like this:
"Speaker: John Doe - 0000-0000-000-0000 owned by Someone Else - 000-000-000-000"

If this shows the questionable text as coming directly from the innocent avatar, then either they are really the cause, or it's a serious exploit. (if the speaker and the owned by data are the same, that indicates the chat came from an avatar and not an object.)

-Atashi

/me still bets on Occams razor..

The Lindens would go monkey poo and stomp all over this, if it was possible.

Thanks, Atashi. Tested it in a preliminary way, and it works as advertised, and looks like it will be very helpful. You just may have nipped a significant griefer problem in the bud. : D

i am so glad someone is on this!

it has happened to me twice at orientation island public

it is embarrassing and hard to explain to people that it is not you doing it

impossible to AR as there is no way to tell who is doing it

Perhaps SLProxy would work.

Thumbs up for an intelligent and helpful response!

QFT!

these type of 'spoofing' devices have been around for a while -- almost a year, at least.

you can probably still buy a few at slexchange, too. but of course, the scripts are no-mod, so you can't see HOW they do the white text.

you can also add a check for agent size into the detector script. llGetAgentSize(key id) will return ZERO_VECTOR if the object is not an agent (avatar). that way, it will only report when objects are speaking.

default {
on_rez(integer n) {
llResetScript();
}
state_entry() {
llListen(0, "", NULL_KEY, "");
}
listen(integer c, string n, key i, string m) 
{
  if(llGetAgentSize(i) != ZERO_VECTOR) return; //--weed out actual avatars speaking

key ownedByKey = llGetOwnerKey(i);
if((ownedByKey==llGetOwner())) return;
string ownedByName = llKey2Name(ownedByKey);
llOwnerSay("Speaker: " + n + " - " + (string)i + " owned by " + ownedByName + " - " + (string)ownedByKey);
}
}

(syntax not tested)

Got a link to one? A (quick) search didn't find anything that advertised that it could do white text and, appologies to Har, I still don't believe it's possible.

to spoof in -white text- with im packet injection, you just add a 'what i say\nPhillip Linden: Hai Guys' to your normal chatter.

The person who is doing the spoofing would have to say something on the line directly above the spoofed chat. if they are using this method its fairly easy to see who is doing it, just look up 1 line of chat

FYI, using SLProxy you can do blue system text. At least, you could a couple weeks ago.