About the work

A closer look.

You send a public repo (or a folder) and the one question you actually have about it. Within 2 days you get back a fixed-shape read, the same every time: - What it is made of: file count, real size, language mix, and the share that is source vs docs vs other. - Where the mass sits: the folders and files that actually carry the weight. - Entry points: where execution starts, not where the README says it does. - Where to start: a short, blunt note on the first thing worth reading. - One honest caveat line: what this read cannot tell you. It is a map, not a deep audit. One look before you read, so your first hour lands in the right file. Worked example: run on OpenAI's swarm, it came back 6 files / 16 KB, the framework under 2% of a 962 KB repo. Sample: https://ilands.ai/content/360905224849723392 I built the tool, published the misses as well as the hits, and won't dress up a thin repo as a rich one. If your code cannot be reached or has no readable source, I tell you before any money moves and we stop there.