Local session-based tracking
CodeBurn reads local session files from supported tools and turns them into token and cost reports without uploading data.
CodeBurn
CodeBurn is a local-first tracker for AI coding token usage and cost across tools like Claude Code, Cursor, and OpenAI Codex CLI. It shows spend by model, project, task, and time range without uploading session data.
What CodeBurn does
CodeBurn is a local-first tool for tracking AI coding token usage and cost across multiple agentic coding tools. It is positioned as a privacy-first alternative to ccusage and is described as free, open source, and 100% local in the supplied page text.
The product reads session files already on your machine and converts them into itemized views of spend by model, project, task, and time range. The site highlights support for Claude Code, Cursor, OpenAI Codex CLI, Copilot, Gemini, and a longer list of tools, with views available in a desktop menu bar app, a GNOME top bar app, and a terminal command via npx codeburn.
Core capabilities
Multi-level spend breakdowns
The product breaks spend down by model, project, and task, so users can see where usage comes from instead of only seeing a total.
Model and task analysis
The Claude Code page notes per-model views such as Opus, Sonnet, and Haiku, plus task views for coding, delegation, exploration, and debugging.
Cross-tool aggregation
The homepage and tool-specific pages describe combined reporting across tools including Claude Code, Cursor, Codex, Copilot, Gemini, and others.
Desktop and terminal access
The homepage describes live spend in the menu bar or GNOME top bar, with a terminal option available through npx codeburn.
Usage efficiency and trend views
The homepage says CodeBurn can show month-to-date totals, projected month spend, cache hit and one-shot rates, and session or pull-request attribution where available.
Common ways people use CodeBurn
Review Claude Code spend
Track Claude Code usage by model and task, including cache pricing and API-equivalent value, to understand whether a subscription is paying off.
See one total across tools
Combine Cursor, Codex, Copilot, and other tools into a single view so you can compare usage across your AI coding stack.
Monitor spend as you work
Watch live spend in the menu bar or top bar during the day so you can notice costs before the end of the month.
Attribute spend to specific work
Inspect project-, branch-, or session-level totals to understand which repository or work item consumed the tokens.
Analyze usage efficiency
Use efficiency views such as one-shot rate, cache hits, retries, and projected month total to assess how your workflow is changing over time.
Pros and Cons
Pros
- Runs locally and reads session files already on the device.
- Supports multiple AI coding tools in one place instead of treating each one separately.
- Breaks usage down by model, project, task, and time range for more detailed cost review.
- Offers both a desktop menu bar/top bar view and a terminal command for different workflows.
- Includes efficiency-oriented views such as cache hits, one-shot rates, and projected month spend.
Cons
- The supplied pages do not provide a pricing or plan structure beyond describing the product as free.
- Some capability details are described at a high level on the homepage, but not all supported tools and workflow features are documented equally deeply in the provided sources.
FAQ
What does CodeBurn track?
CodeBurn runs locally and reads session files from tools such as Claude Code, Cursor, and Codex. It then shows token usage and estimated cost by model, project, task, and time range.
Is CodeBurn free to use?
The source materials describe CodeBurn as free, local-first, and open source. They do not provide a detailed paid-plan breakdown on the pages provided.
Which platforms does CodeBurn support?
The pages describe support for macOS, Windows, and Linux. The homepage also notes a menu bar app on macOS and a GNOME top bar view on Linux.
Can CodeBurn combine usage from multiple AI coding tools?
Yes. The Cursor page says CodeBurn can combine Cursor usage with Claude Code, Codex, Copilot, Gemini, and more so you can see one number for all of it.
Quick Facts
- Category
- Developer Tool
- Primary use
- AI coding cost and token usage tracking
- Supported tools mentioned
- Claude Code, Cursor, OpenAI Codex CLI, Copilot, Gemini, and more
- Platforms mentioned
- macOS, Windows, Linux
- Access methods
- Menu bar app, GNOME top bar, terminal command (npx codeburn)
- Source domain
- codeburn.app
Alternative a CodeBurn
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
CreateOS Sandbox
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
hob
hob is an independent workspace for coding agents that keeps agent sessions, terminals, history, and follow-up work organized around the tools and providers you already use. It is aimed at developers who want local control over routing, history, and workspace structure rather than a bundled model stack.
Manta AI
Manta AI is an autonomous web app testing tool for teams that want to map application behavior, catch regressions, and generate tests without writing scripts or maintaining selectors. It works from a URL and supports plain-English test flows, run results with screenshots, and scheduled or deployment-triggered checks.
Redline
Redline is a budgeting tool for Claude Code that paces sessions to stay within time, token, cost, or plan-percentage limits. It uses Claude Code’s native hooks and statusline to help sessions finish with a usable result instead of stopping abruptly.
SonOf
SonOf connects to your repo and PM tool, audits the codebase and surrounding product context, and turns approved work into shipped tickets with senior engineering review. It is aimed at founders and engineering leaders who need backlog help without hiring a full team immediately.