The standard answer is "clone it and run cloc." That's still the right call for a precise, language-by-language audit — but if you just want a fast sense of scale before deciding whether to dig in further, there's a much lighter option.
"How many lines of code does this repo have" usually means one of two things. Sometimes it's a precise question — you're doing a formal audit, generating a report, or need an exact per-language breakdown, in which case a dedicated command-line tool run against a full local clone is genuinely the right instrument. Other times it's a rough, exploratory question — "is this a 500-line script or a 50,000-line application?" — asked before you've committed to cloning anything at all. RepoKit is built for the second case.
If you need a precise, defensible number — for a report, a licensing review, or a formal codebase assessment — clone the repo and run cloc or tokei, which categorize by language and separate code from comments and blank lines. RepoKit's estimate is for fast orientation, not compliance documentation.
GitHub already shows a colored language bar on every repo, but it's measuring bytes per language, not lines — a file full of long lines and a file full of short lines can show identical byte proportions while representing very different amounts of actual code. Line count is a closer proxy for "how much did someone actually write here," which is usually the real question behind "how big is this codebase."
Repocloc or tokei locally gives you precise, categorized line counts (broken down by language, comments vs. code, etc). That's the right choice for a serious one-time audit. RepoKit trades that precision for zero setup and instant, repo-wide coverage without leaving your browser.No cloning, no command line — just open the repo and read the numbers.
Add RepoKit to Chrome — it's free