"imidscript_source" should be "timidscript_source". (x4)
"TinyEye" should be "TinEye".

            panel = CreatePanelControlIcon("SauceNAO", "sourceSauce", "http://saucenao.com/search.php?db=999&url=" + data.imgSrc + "&imidscript_source");
            linkPanel.appendChild(panel);
            panel = CreatePanelControlIcon("ImgOps", "sourceImgOps", "http://imgops.com/" + data.imgSrc.replace("http://", "") + "?imidscript_source");
            linkPanel.appendChild(panel);
            panel = CreatePanelControlIcon("Google", "sourceGoogle", " http://google.com/searchbyimage?hl=en&site=search&image_url=" + data.imgSrc + "&imidscript_source");
            linkPanel.appendChild(panel);
            panel = CreatePanelControlIcon("TinyEye", "sourceTinyEye", "http://www.tineye.com/parse?url=" + data.imgSrc + "&imidscript_source");
            linkPanel.appendChild(panel);

Re: @TimidScript:

The fix is working for me, but should the illustration title be displaying above the three grey lines indicating a manga image? (Thumbnail, Title, Lines, Manga Links, Artist Name)

Unless it would take a significant amount of coding to change, I would think it should be displaying below the lines. It's not a big issue though.

Everything seems to be fine. I would have changed the gm_xhr in GIviewer as well, but neither gm_xhr or xhr want to call onload at all. It's beyond my skills I think, and the only thing it did was display the artist's avatar in the linkbox with a link to their profile.

There's a couple other small things but I'll put them in the proper tickets.

Also, here's the manga viewer for good measure.

http://pastebin.com/raw.php?i=41UE1LJn

An update: I replaced each GM_xml and everything's back to normal. Either there's something not functioning as intended in GM_xml itself or there's some setting you're supposed to toggle, I guess.
Hope this helps.

http://pastebin.com/raw.php?i=4ryzBjUa

It's a good thought, but nothing changed. Yes, method #3 works.

Okay. I went in the script and changed the first http request to this:

           GM_xmlhttpRequest({
               url: "http://www.pixiv.net/member_illust.php?mode=medium&illust_id=" + id,
               method: "GET",
               timeout: 15000,
               //headers: { "User-agent": navigator.userAgent, "Accept": "text/html", Referer: "http://www.pixiv.net" },
               headers: { "User-agent": navigator.userAgent, "Accept": "text/html" },
               onload: function (response)
               {
                   alert("GM_xml:" + response.responseText);
               }
           });
           
           var oReq = new XMLHttpRequest();
           oReq.open("GET", "http://www.pixiv.net/member_illust.php?mode=medium&illust_id=" + id, true);
           oReq.responseType = "text";
           oReq.timeout = 15000;
           oReq.onload = function (e)
           {
               alert("xml:" + oReq.response);
           };
           oReq.send();

All GM_xml pages returned contain pixiv.user.loggedIn = false; all xml pages returned contain pixiv.user.loggedIn = true. Commenting out either (or both) header lines doesn't change that.

I don't know javascript but I know the basics of scripting. It took me a while but eventually I realized that GM_XmlHttpRequest is returning non-logged-in pages. Sure enough, NextPageMethod #1 (GM_XmlHttpRequest) fails, but NextPageMethod #2 (XmlHttpRequest) works. Something must be different on the authentication or cookie backend, I guess?

At least it looks like a quick fix. Adding a variable to change the three GM_XmlHttpRequests in getDataHTML to XmlHttpRequests should work perfectly and only take a couple minutes for someone who actually knows javascript.

Thanks for your work, by the way. I used this all the time back when in was on USO and didn't want to browse pixiv without it.

Pixiv Lazy Plus works correctly.

When logging into pixiv via Google, some things break.

The illustration linker completely breaks; fails to get any metadata and never displays anything. When using paging more pages can be scrolled to, but the message still reads "Getting metadata for [20] illustrations..." instead of increasing to 40 as expected.

On the illustration page itself, again, responses and bookmarks are listed as 0. On manga illustration pages, all of the direct links displayed link to a thumbnail of the artist's avatar. This happens on the manga page itself too when using your manga viewer script.

Everything else seems to work correctly. In particular, the thumbnails display correctly on a manga illustration page.

This is interesting. I haven't used it much but I can see where it is useful.

One thing I would like to see supported if possible is 90 degree image rotation... I used to do that a lot. I think all the old plugins I used are outdated now though. Having rotate left and rotate right in addition to the image sizing would make this one of my main scripts I think.