Xcode.life used to offer similar utilities, but I believe that went into abeyance when they began focusing on the sale of medical and wellness reports. Some Googling might turn up other options, including some for-fee options like, perhaps, DNAgedcom.
But it really isn't all that difficult to convert the results yourself. One important note here is that MyHeritage moved to a version of the GSA v2 chip as of last April, so step-by-step instructions really can't be provided because they'll differ with the format.
Both are just plain text files, though, and both still use the (now outdated, but it's still what our microarray tests use) human genome reference assembly 37, also referred to as hg19 under the U.C. Santa Cruz nomenclature. The caveat is that the naming of the tested SNPs--specifically the rsIDs, or "reference sequence IDs"--is not entirely consistent, even though the dbSNP database is supposed to be the arbiter.
If you want to dive in yourself, I'd recommend using something like MS Excel to open and sort the MyHeritage file by chromosome number as the first criterion, and secondarily by position, or locus.
Throw away anything that might show as being chromosome "0" and, unless you want to undertake some significant trial-and-error (create a file; try to upload it; fails; modify the file; try again), throw away anything that does not have a name beginning with "rs" as the first element of the row. You'll likely lose several hundred markers that way, but it should produce a clean file.
Here's an example from Roberta Estes of a 23andMe file format: https://dnaexplained.files.wordpress.com/2013/11/build-37.png.
The lines commented out at the top with an asterisk will still be evaluated by machine-reading sources. The first line and the one heading the actual data ("# rsid chromosome position genotype") in particular. As long you include those, it should work correctly.
The data is tab-delimited: each field in each record (row) separated by the "tab" character. Note that the actual nucleotide values are combined together under the "genotype" column; the letters aren't separated. Your MyHeritage data should be laid out the same way, but the headings are probably going to be "RSID CHROMOSOME POSITION RESULT." Change the headings to the 23andMe format, including the initial "#" and it should be good to go. There is no special indicator for the end of the data; it just ends with the last row.
Save it as a tab-delimited file in Excel (.csv), rename it to .txt, and compress it to a .zip file. The filename takes the form: genome_Firstname_Lastname_20192345678901.txt, zipped as genome_Firstname_Lastname_20192345678901.zip; change the Firstname and Lastname appropriately, and choose a random series of numerals after the "2019."
That should do it. Others have mentioned GEDmatch, and it's worthwhile noting that they now have, as an option under the Tier 1 subscription, the ability to combine results from multiple companies into a single "super-kit." If wanting genealogical comparisons, that's a good way to go. My kit there now has almost a million SNPs rather than any single kit's ~650,000, and the added density of SNPs improves matching. I know of others who have gone over the 1.2 million mark.
Have fun!
