This is simply a set of small tools for use with SCI16, some of them mostly for SCI11+, made by Kawa.
Very simple deal, you give it an image file (PNG preferred) with a particular layout and it'll give you a Sierra SCI Font resource. If you simply drag and drop the file onto the application icon, it'll assume you want the same filename for the output but with ".fon" at the end instead. The first character cell found is used to set the font's line height and does not itself appear in the result. Duplicates (for example those little single-dot placeholders) are optimized away.
A cell is any rectangular area that's only pure black or white fully surrounded by anything else.
You can also give it a Sierra FON file and get a PNG in the correct format.
Even simpler deal. You run it, it produces a 999.voc resource, a list of kernel call names. If you provide a file that's simply one bare name on each line (by dropping it like with Pic2SciFont, for example), it'll use that instead of a hardcoded list.
If there's a 997.txt and no 997.voc file, or the former is more recently changed than the latter, this one'll create a 997.voc resource, a list of selectors, based on the text file. The exact opposite also holds. If you pass a file name, it'll turn a .txt file into a similarly-named .voc file and vice versa. The interesting bit is that you can use //<num> comment lines to skip ahead, which is quite a space saver when you consider -objID- is #4096.
Given a message resource file in text format as exported by SV, this'll compile it back into a valid resource file. But there's a twist: if the first line is !utf8, the text will be read and stored as such. It also accepts // comments and blank lines. Blank lines that are part of a message start with five tabs and won't be skipped.
AutoSave\05 Interval can thus be rewritten as AutoSave™ Interval, where the trademark is stored as-is as a three-byte sequence instead of a three-byte escape. UTF-8 message resources have a little marker at the end, ignored by the interpreter.
Again, you can also provide a message resource file, and get a properly-formatted text file back, including the !utf8 line if the input calls for it.
If there are .sh files available, these will be used to clarify the noun, verb, condition, and talker values. Talker values are taken from talkers.sh, verbs from verbs.sh, nouns and conditions from <basename>.sh. This uses a very simple parser that works on SCI Companion's files so don't muck them up.
Given a .seq file, this will play the file in a window -- click the close button or press Esc to exit. Given the first of an image sequence (something0001.png or something-1.png, who cares), this will instead take that image sequence and make a .seq file from it (something.seq). There are more options on the command line, such as playing back double-sized with -d, aspect-corrected with -a, exporting frames as they are played with -e, or examining which parts are actually stored for each frame with -x.
Given a .s file with only data, this will happily produce a raw binary. It supports only include, org, equ, db, dw, and dd commands, no actual machine code. Given a single file name, it'll put .bin at the end.