Sumit Kumar - Projects

I was writing up documentation for a project and wanted a clean Markdown view of its folder structure to drop straight into a README. tree on Unix gets you 80% there, but it doesn't output Markdown and isn't very customizable - so instead of fighting sed/awk to reformat its output, I just wrote the tool I actually wanted, in Go.

DirScanner scans a directory and writes its structure out as a proper Markdown tree:

├── .dirignore
├── file2.txt
└── sumit
    ├── file1.txt
    └── sumit-1
        ├── kafka.txt
        └── sumit-2
            ├── file.txt
            └── sumit-3
                └── file3.txt

Beyond the basic tree dump, it picked up a handful of features that made it genuinely useful instead of a toy:

• Custom connector styles - swap the branch characters (+--, `--, whatever fits your doc style)
• Exclude files/dirs - skip node_modules, specific extensions, or anything else that clutters a real tree
• Depth limiting - cap how deep it recurses instead of dumping an entire monorepo
• .dirignore support - works exactly like .gitignore, so you don't have to pass exclude flags every time

Install it directly with:

go install github.com/SumitKumar-17/DirScanner@latest

and run it with:

dirscanner ./myproject tree.md --exclude ".txt" --exclude "node_modules" --depth 2

Small tool, but a genuinely good excuse to actually learn Go's filesystem APIs and CLI flag parsing instead of just reading about them - and I use it often enough on other projects that it's paid for the time it took to build.