# getagentseal/codeburn

Free, local tool to track AI coding token usage and cost across 37 tools and agents (Claude Code, Cursor, Codex, Gemini and more), by model, project, and task. npx codeburn

Repository: https://github.com/getagentseal/codeburn
Canonical: https://ross.abutalabs.com/products/codeburn
Homepage: https://codeburn.app/
Language: TypeScript
License: MIT
License Family: permissive
Topics: ai-coding, claude-code, cli, codex, cost-tracking, developer-tools, observability, terminal-ui, token-usage, cursor-ide, menubar, web-dashboard
Last push: 2026-08-25T18:32:46+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 99, longevity 10
- inputs: {"age_days": 142, "days_push": 8, "days_rel": 8, "gap_med": 0.0, "n_releases_24m": 57}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 9676, forks 769 (observed 2026-08-28T04:10:36.663856+00:00)

## What it is
CodeBurn is a free, local-first tool that tracks AI coding token usage and cost across 41 AI coding tools (Claude Code, Cursor, Codex, Gemini CLI, and more) by model, project, task, and session. It ships as an npx-runnable terminal dashboard plus a native macOS menu bar app, GNOME extension, and desktop app, reading session files directly from disk with nothing leaving the machine.

## Use cases
- track how much my AI coding assistant costs per day
- see token spend by model across Claude Code and Cursor
- find which projects burn the most AI tokens
- reduce AI coding costs by finding token waste
- monitor AI spend in the macOS menu bar
- compare cost efficiency of different AI coding models
- attribute AI spend to git branches and pull requests

## When to choose
- you use multiple AI coding tools and want one unified local cost ledger
- you want spend visibility without proxies, API keys, or cloud services
- you need per-project, per-model, per-task breakdowns with waste-detection fixes

## When to avoid
- you need team-wide or org-level centralized spend reporting
- you want to track non-coding LLM API usage outside coding tools
- you require real-time billing data from provider APIs rather than local session files

## Facets
- artifact type: cli-tool
- maturity: active
- function: monitoring, analytics, cli, terminal-ui, developer-tools
- domain: developer-tools, analytics, large-language-models
- platform: windows, cli
- tags: token-usage, cost-tracking, ai-coding, observability, menubar-app, local-only, litellm-pricing, claude-code, cursor, codex, command-line, macos, linux, nodejs, desktop

## Member repositories
- getagentseal/codeburn (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:36.663856+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:21:17.501526+00:00, confidence not recorded.
  - readme: https://github.com/getagentseal/codeburn (fetched 2026-08-28T04:10:36.663856+00:00, sha a2461285442a)
  - homepage: https://codeburn.app/ (fetched 2026-08-29T08:21:27.426522+00:00, sha 3ecaf8158e0c)
  - site_page: https://codeburn.app/docs (fetched 2026-08-29T08:21:27.429930+00:00, sha d66586ba9225)
  - site_page: https://codeburn.app/docs/optimize (fetched 2026-08-29T08:21:27.439791+00:00, sha 8ea57396965e)
  - site_page: https://codeburn.app/docs/providers (fetched 2026-08-29T08:21:27.442486+00:00, sha 6883f4c99ef4)
  - registry_npm: https://registry.npmjs.org/codeburn (fetched 2026-08-29T08:21:27.444572+00:00, sha 34023414907f)
  - site_page: https://codeburn.app/changelog (fetched 2026-08-29T08:21:27.432168+00:00, sha d3cc848d0b93)
- Data as of 2026-08-30T08:39:29.467469+00:00.