![\"Codex](\"https://raw.githubusercontent.com/getagentseal/codeburn/main/assets/open-source-recipient.png\")
\n ![\"CodeBurn](\"https://api.producthunt.com/widgets/embed-image/v1/featured.svg?post_id=1220451&theme=dark&t=1786459783669\")
\n | \n \n | \n \n \n | \n
\n ![\"CodeBurn\"](\"https://raw.githubusercontent.com/getagentseal/codeburn/main/assets/providers.png\")
\n
See where your AI spend goes.
\n\n\n ![\"npm](\"https://img.shields.io/npm/v/codeburn.svg?color=F97316\")
\n ![\"total](\"https://img.shields.io/npm/dt/codeburn.svg?color=F97316\")
\n ![\"license\"](\"https://img.shields.io/npm/l/codeburn.svg?color=F97316\")
\n ![\"node](\"https://img.shields.io/badge/node-%3E%3D22-F97316.svg\"){kind=icon}
\n ![\"Discord\"](\"https://img.shields.io/badge/discord-join-F97316?logo=discord&logoColor=white\"){kind=icon}
![\"Follow](\"https://img.shields.io/badge/%40__codeburn-F97316?logo=x&logoColor=white\"){kind=icon}
![\"Sponsor\"](\"https://img.shields.io/badge/sponsor-♥-F97316?logo=github\"){kind=icon}
If CodeBurn shows you something your bill never did, star the repo so other developers find it, and consider sponsoring to keep 41 integrations honest.
\n\n| \n Desktop \n ![\"CodeBurn](\"https://raw.githubusercontent.com/getagentseal/codeburn/main/assets/desktop.jpg\") \n ![\"Download](\"https://img.shields.io/badge/macOS-Apple_Silicon-F97316?logo=apple&logoColor=white\"){kind=icon} \n ![\"Download](\"https://img.shields.io/badge/macOS-Intel-F97316?logo=apple&logoColor=white\"){kind=icon} \n ![\"Get](\"https://img.shields.io/badge/Windows-Microsoft_Store-F97316?logo=microsoft&logoColor=white\"){kind=icon} \n ![\"Download](\"https://img.shields.io/badge/Linux-.deb-F97316?logo=debian&logoColor=white\"){kind=icon} \n ![\"Download](\"https://img.shields.io/badge/Linux-.rpm-F97316?logo=redhat&logoColor=white\"){kind=icon} \n ![\"Download](\"https://img.shields.io/badge/Linux-AppImage-F97316?logo=linux&logoColor=white\"){kind=icon} \n | \n \n Web \n ![\"CodeBurn](\"https://raw.githubusercontent.com/getagentseal/codeburn/main/assets/web.jpg\") \n npx codeburn web\n | \n
| \n Terminal \n ![\"CodeBurn](\"https://raw.githubusercontent.com/getagentseal/codeburn/main/assets/dashboard.jpg\") \n npx codeburn\n | \n \n Menubar \n ![\"CodeBurn](\"https://raw.githubusercontent.com/getagentseal/codeburn/main/assets/menubar-app.jpg\") \n codeburn menubar\n \n ![\"Download](\"https://img.shields.io/badge/Windows-Tray_app_.msi_(preview)-F97316?logo=windows&logoColor=white\"){kind=icon} \n | \n
| \n Capacity Dock · native macOS menubar \n ![\"CodeBurn](\"https://raw.githubusercontent.com/getagentseal/codeburn/main/assets/capacity-dock.jpg\") \n codeburn menubar\n | \n
One source of truth: every surface reads the session files already on your disk.
\n\n**CodeBurn is a free, open-source, local-first tool that tracks AI coding token usage and cost across 41 tools and agents (Claude Code, Cursor, Codex, Gemini, Grok and more), broken down by model, project, and task.**\n\nYou pay for Claude, Codex, Cursor, and a stack of other AI tools. The bill tells you the total. It never tells you that half of it went to conversation instead of code, or that an expensive model burned your budget on work a cheaper one would have one-shot.\n\nCodeBurn does. It reads the session files your tools already write to disk and breaks down every token and dollar by **task, model, tool, and project**, across **41 AI tools**.\n\nEverything runs locally. No wrapper, no proxy, no API keys, nothing leaves your machine. Pricing comes from [LiteLLM](https://github.com/BerriAI/litellm), refreshed daily. The optional desktop app can send an anonymous, bucketed usage report if you say yes to it; see [Telemetry](#telemetry).\n\n\n Quick start ·\n Find waste ·\n Apply fixes ·\n Guard ·\n Compare models ·\n Track what shipped ·\n MCP ·\n Supported tools ·\n Commands ·\n Features ·\n How it reads data ·\n Telemetry\n
\n\n## Quick start\n\n**Run it instantly**, no install needed:\n\n```bash\nnpx codeburn\n```\n\nThat opens the interactive dashboard (today by default, or the last 7 days when today has no usage yet). Arrow keys switch periods, `q` quits. That is the 30-second version. You now know where your AI budget goes.\n\n**Install it** for a permanent `codeburn` command:\n\n```bash\nnpm install -g codeburn\n```\n\nAlso runs via `bunx codeburn` or `pnpm dlx codeburn`, or `brew install codeburn` on macOS.\n\n**Menu bar app** for macOS, with your spend always in the menu bar:\n\n```bash\ncodeburn menubar\n```\n\nThe same command installs the tray app on Windows; see [Windows](#windows). On Linux, a GNOME Shell extension gives it in the top panel; see [Linux (GNOME)](#linux-gnome).\n\nRequires **Node.js 22.13+** and at least one supported tool with session data on disk. For Cursor and OpenCode, `better-sqlite3` installs automatically.\n\n## Your month at a glance\n\n```bash\ncodeburn overview # this month, clean tables\ncodeburn overview --no-color # plain text, ready to paste\ncodeburn overview --from 2026-06-01 --to 2026-06-15 # any date range\ncodeburn overview -p all # last 6 months\ncodeburn overview -p lifetime # full history (uncapped)\ncodeburn overview --provider claude # one tool only\n```\n\n`codeburn overview` prints a copy-pasteable summary of where your AI spend went: totals (cost, tokens, cache hit), a breakdown by tool and by top model, your highest-value days, top projects, a per-day table, and activity and tool usage. Pipe it anywhere (into `pbcopy`, a PR, Slack, or a tweet); color drops automatically when the output is not a terminal, or pass `--no-color`.\n\n```text\nCodeBurn June 2026\n\nTotals\n Cost $2,795.10\n Tokens 3.49B in 23.9M / out 20.2M / cache-w 72.5M / cache-r 3.38B\n Calls 14,755 sessions 753\n Cache hit 99.3%\n\nBy tool\n┌──────────┬───────────┬────────┬───────┐\n│ Tool │ Cost │ Tokens │ Share │\n├──────────┼───────────┼────────┼───────┤\n│ claude │ $2,662.37 │ 3.34B │ 95% │\n│ codex │ $119.12 │ 128.1M │ 4% │\n└──────────┴───────────┴────────┴───────┘\n\n(plus Top models, Highest-value days, Top projects, a per-day table, By activity, and Tools)\n```\n\n## Find and fix waste\n\n```bash\ncodeburn optimize # scan the last 30 days\ncodeburn optimize -p today # today only\ncodeburn optimize -p week # last 7 days\ncodeburn optimize --provider claude # restrict to one provider\ncodeburn optimize --format json # setup health + findings as JSON\n```\n\n`codeburn optimize` scans your sessions and your `~/.claude/` setup for waste patterns:\n\nFor Claude Code, the optimize session count, the per-session findings, coaching,\nand model-default recommendations use user-started (main) sessions. Subagent\nsidechain transcripts are excluded from that population because their delegated\ncontext and delivery behavior are structurally different, and so is the re-read\nfinding, since a subagent starts on a fresh context. Findings about how Claude\nuses tools (junk reads, read:edit ratio) and every spend, MCP, and\nconfiguration-overhead finding keep counting them.\n\n- Files Claude re-reads across sessions (same content, same context, over and over)\n- Low Read:Edit ratio (editing without reading leads to retries and wasted tokens)\n- Wasted bash output (uncapped `BASH_MAX_OUTPUT_LENGTH`, trailing noise)\n- Unused MCP servers still paying their tool-schema overhead every session\n- Ghost agents, skills, and slash commands defined in `~/.claude/` but never invoked\n- Bloated `CLAUDE.md` files (with `@-import` expansion counted)\n- Cache creation overhead and junk directory reads\n- Context-heavy sessions where effective input/cache tokens swamp output\n- Possibly low-worth expensive sessions with no edit turns or repeated retries\n when no `git`/`gh` delivery command is observed\n\nFindings are grouped into three classes: **Fix now** (CodeBurn can apply it for you), **Habits**\n(you change how you drive the next session), and **FYI** (informational, the cost may be justified).\nEach one says whether its savings number is `measured` from provider-counted usage or `estimated`\nfrom a model. See [docs/optimize.md](docs/optimize.md) for what is scanned, what `--apply` may write,\nand how to read the health grade.\n\nEach finding shows the estimated token and dollar savings plus a ready-to-paste fix: a `CLAUDE.md` line, an environment variable, or a `mv` command to archive unused items. Findings are ranked by urgency (impact weighted against observed waste) and rolled up into an A to F setup health grade. Repeat runs classify each finding as new, improving, or resolved against a 48-hour recent window.\n\nYou can also open it inline from the dashboard: press `o` when a finding count appears in the status bar, `b` to return.\n\n## Apply fixes, undo anytime\n\n```bash\ncodeburn optimize --apply # review and apply fixes interactively\ncodeburn optimize --apply --dry-run # print the plan, change nothing\ncodeburn optimize --apply --yes # apply every appliable fix without prompting\ncodeburn act list # every change CodeBurn has made\ncodeburn act undo --last # roll the most recent change back\ncodeburn act report # realized vs estimated savings\ncodeburn optimize --auto-revert # undo the applied fixes that measured no reduction\n```\n\n`codeburn optimize` finds the waste; `--apply` fixes the config-class findings for you: settings values, environment variables, archiving unused agents and skills. Every change is backed up and journaled before it lands. `codeburn act list` shows the history and `codeburn act undo\n ![\"Claude](\"assets/providers/claude.jpg\")
\n ![\"Cline\"](\"assets/providers/cline.svg\")
\n ![\"CodeWhale\"](\"assets/providers/codewhale.svg\")
\n ![\"Codex](\"assets/providers/codex.png\")
\n ![\"Cursor\"](\"assets/providers/cursor.jpg\")
\n ![\"cursor-agent\"](\"assets/providers/cursor-agent.jpg\")
\n ![\"Devin\"](\"assets/providers/devin.png\")
\n ![\"Forge\"](\"assets/providers/forge.png\")
\n ![\"Gemini](\"assets/providers/gemini.png\")
\n ![\"Mistral](\"assets/providers/mistral-vibe.svg\")
\n ![\"GitHub](\"assets/providers/copilot.jpg\")
\n ![\"IBM](\"assets/providers/ibm-bob.svg\")
\n ![\"Kiro\"](\"assets/providers/kiro.png\")
\n ![\"OpenCode\"](\"assets/providers/opencode.png\")
\n ![\"OpenClaw\"](\"assets/providers/openclaw.jpg\")
\n ![\"Pi\"](\"assets/providers/pi.png\")
\n ![\"OMP](\"assets/providers/omp.svg\")
\n ![\"Droid\"](\"assets/providers/droid.png\")
\n ![\"Roo](\"assets/providers/roo-code.png\")
\n ![\"KiloCode\"](\"assets/providers/kilo-code.png\")
\n ![\"Qwen\"](\"assets/providers/qwen.png\")
\n ![\"Kimi](\"assets/providers/kimi.svg\")
\n LingTai TUI\n ![\"Goose\"](\"assets/providers/goose.png\")
\n ![\"Antigravity\"](\"assets/providers/antigravity.png\")
\n ![\"Crush\"](\"assets/providers/crush.png\")
\n ![\"Warp\"](\"assets/providers/warp.jpg\")
\n ![\"Mux](\"assets/providers/mux.png\")
\n ![\"Vercel](\"assets/providers/vercel-gateway.png\")
\n ![\"Zerostack\"](\"assets/providers/zerostack.png\")
\n ![\"Grok](\"assets/providers/grok.png\")
\n ![\"ZCode\"](\"assets/providers/zcode.jpg\")
\n ![\"Zed\"](\"assets/providers/zed.jpg\")
\n ![\"Hermes](\"assets/providers/hermes.png\")
\n