GitHub hosts HunyuanVideo, Tencent's open-source framework for large-scale video generation models, enabling AI-driven video creation.
Track your Claude Code usage
€3Starting price
0Popularity
About Track your Claude Code usage
This tool captures full execution traces and token spend for coding agents such as Claude Code, Codex, Copilot, Cursor, and Pi. It records detailed spans, duration, input/output tokens, cache operations, and model usage across sessions. The interface displays aggregated cost metrics, performance comparisons against percentiles, and conversation summaries. Users can filter traces by model, duration, or cost to analyze agent behavior patterns. The tool also provides visibility into reasoning effort, service origins, and error contexts within agent workflows. It supports tracking of both billed and theoretical token costs under bundled plans.
Key features
- Trace every agent step
- Monitor token spend and latency
- Compare performance against percentiles
- Filter traces by model or duration
- Track input/output tokens and cache operations
- View conversation summaries and error contexts
- Support for multiple coding agents
- GitHub sync for prompt management
Use cases
- Debugging agent workflows and identifying performance bottlenecks
- Tracking token usage and cost for budgeting
- Analyzing agent behavior across different models and sessions
Pros
- Full trace history for coding agents
- Token spend tracking and cost analysis
- Performance percentile comparisons
- Multi-model support (Claude, Copilot, Cursor, etc.)
- GitHub integration for prompt management
Cons
- No free tier beyond 50k events/month
- Self-hosting requires setup
- Enterprise features require custom pricing
Frequently asked questions about Track your Claude Code usage
What is Track your Claude Code usage and what does it do?
Track your Claude Code usage is a feature within the LangWatch platform that captures full execution traces and token spend for coding agents such as Claude Code, Codex, Copilot, Cursor, and Pi. It records detailed spans, duration, input/output tokens, cache operations, and model usage across sessions.
Who should use Track your Claude Code usage?
This tool is designed for developers, engineering teams, and organizations that use AI coding agents and need visibility into their usage, performance, and costs. It helps users analyze agent behavior patterns and optimize their workflows.
How does Track your Claude Code usage track token spend?
It tracks token spend by recording input, output, cache read, and cache write operations for each session. The tool displays both billed and theoretical token costs, allowing users to compare and analyze spending patterns under bundled plans.
Can I filter and analyze the traces captured by this tool?
Yes, users can filter traces by model, duration, or cost to analyze agent behavior patterns. The interface provides aggregated cost metrics, performance comparisons against percentiles, and conversation summaries for deeper insights.
What kind of visibility does Track your Claude Code usage provide into agent workflows?
The tool offers visibility into reasoning effort, service origins, error contexts, and detailed spans within agent workflows. It also includes features like waterfall, flame, topology, and sequence views to help users understand agent execution.
How do I get started with Track your Claude Code usage?
Users can start tracking right away by running a command such as 'npx langwatch claude' in their terminal. The tool integrates with coding agents and begins capturing traces and token usage immediately.
