# Install codeburn with Homebrew

See where your AI coding tokens go - by task, tool, model, and project. Version 0.9.19 via Homebrew; verified 2026-07-21.

## Install

```sh
sudo av install brew:codeburn
```

Additional install commands:

### macOS

- Homebrew (100%):

```sh
brew install codeburn
```

  Evidence: local Homebrew formula metadata

## Package facts

- **Package key:** brew:codeburn
- **Package manager:** Homebrew
- **Version:** 0.9.19
- **Source summary:** See where your AI coding tokens go - by task, tool, model, and project
- **Homepage:** <https://codeburn.app/>
- **Last updated:** 2026-07-21T19:01:49Z
- **Generated:** 2026-08-03T19:37:03+00:00

## Executables

- codeburn (alias)

## Install behavior

- Bottle: not available

## Freshness

- Page generated: 2026-08-03
- Package-manager version: 0.9.19
## Project history and usage

CodeBurn is a local-first CLI, TUI, web dashboard, and macOS menubar tool for analyzing token usage and cost across AI coding tools. It reads session files already written on disk and reports spend by tool, model, task, and project.

### Project history

The GitHub repository was created on 2026-04-13. Its README describes the project as a free, open-source, local tool for tracking AI coding token usage and cost across Claude Code, Cursor, Codex, Gemini, Grok, and other tools.

Early releases moved quickly in April 2026, with the release metadata showing v0.7.0 adding `codeburn optimize` and adjacent macOS menubar releases appearing the same week.

### Adoption history

The README documents `npx codeburn`, global npm installation, Bun/Deno-style runners, and `brew install codeburn` on macOS. The input package facts currently list Homebrew packaging.

Because CodeBurn targets local files created by other AI coding tools, its adoption story is tied to the spread of agentic coding CLIs and editors rather than to one language ecosystem.

### How it is used

Users run `npx codeburn` or `codeburn` to open an interactive dashboard, `codeburn overview` for reports, `codeburn optimize` for waste findings, `codeburn compare` for model comparisons, `codeburn yield` to correlate sessions with git commits, and `codeburn web` for a localhost dashboard.

The README stresses local operation: no wrapper, proxy, or API keys, with pricing sourced from LiteLLM and refreshed daily.

### Why package nerds care

CodeBurn is notable as packaging infrastructure for the AI-coding era: it treats token logs and session files like operational telemetry and ships through npm-style runners and Homebrew. For package nerds, the interesting part is how a Node CLI, local dashboard, native menubar app, and Homebrew formula package one local observability workflow.

### Timeline

- 2026-04-13: GitHub repository created.
- 2026-04-16: v0.7.0 release metadata names `codeburn optimize`.
- 2026-04-18 to 2026-04-19: Multiple macOS menubar releases published.
- 2026: README documents npm, npx, Bun/Deno-style runners, and Homebrew installation.

### Related projects

- The README names Claude Code, Cursor, Codex, Gemini, Grok, and LiteLLM in the supported-tooling story. The macOS menubar app lives in the same repository under `mac/` according to the README.

### Sources

- <https://github.com/getagentseal/codeburn>


## Security Notes

No matching local secret-handling manifest was found for codeburn. Nucleus package metadata is still published here so future coverage has a stable package URL.



## Combined YAML source

View the package source record on GitHub. [combined/codeburn.yml](https://github.com/mxcl/pkgdb/blob/main/combined/codeburn.yml)


## Sources

- pkg.so package database
- curated package history
- pkgdb category and tag curation
- cross-ecosystem install command graph