gmarcais/jellyfish — reverse-engineered prompt

Build me a fast C++ command line tool for DNA sequence analysis, like Jellyfish. It should read FASTA and multi FASTA files, count every k mer of a length I choose, and use multiple CPU threads so it stays fast and does not waste memory on large genomes.

Save the counts in a compact binary file, then give me simple commands to turn that file into readable text and to ask for the count of one specific k mer. I want normal help text, clear error messages, examples I can run, and basic tests with small sample FASTA files so I can trust the results.

Please include build and install instructions for Linux, and keep the code organized so the counting engine can also be used from other programs later. If you need details about accepted Jellyfish style command options, look up the current docs online.