Tokei — Fast Code Statistics for Any Language

What it is

Tokei is a code statistics tool written in Rust that counts lines of code, comments, and blanks across 200+ programming languages. It runs in milliseconds even on large codebases and outputs clean tables showing the breakdown by language and file. It provides fast, accurate metrics for project health reports and AI context.

It is designed for developers, tech leads, and CI pipelines that need quick codebase metrics without the overhead of full static analysis tools.

How it saves time or tokens

The token estimate for this workflow is 2,200 tokens. Running tokei in a project directory gives you an instant snapshot of codebase composition. For AI agents, this provides essential context (language distribution, project size) in a single command rather than manual file counting or guessing.

How to use

1. Install: brew install tokei (macOS) or cargo install tokei (any platform)
2. Run tokei in your project directory
3. View the language-by-language breakdown of code, comments, and blanks

Example

# Install
brew install tokei  # macOS
cargo install tokei  # any platform with Rust

# Count lines in current project
tokei

# Output:
# ===============================================================================
#  Language            Files        Lines         Code     Comments       Blanks
# ===============================================================================
#  TypeScript             45         3420         2810          280          330
#  Python                 12          980          820           60          100
#  JSON                    8          450          450            0            0
# ===============================================================================
#  Total                  65         4850         4080          340          430
# ===============================================================================

# Exclude directories
tokei --exclude node_modules --exclude .git

# Output as JSON for programmatic use
tokei --output json

# Sort by lines of code
tokei --sort code

Related on TokRepo

• AI Tools for Coding -- Developer tools for codebase analysis
• Featured Workflows -- Top-rated workflows on TokRepo

Common pitfalls

• Tokei counts by file extension; files without extensions or with non-standard extensions may be misclassified or skipped
• Generated files (node_modules, build outputs) inflate counts; use --exclude to filter them
• Comment counting accuracy depends on Tokei's language parser; some complex comment syntaxes may be miscounted