agenttrace agenttrace
agenttrace is a local-first terminal-based observability tool that parses logs from various AI coding agents to track session history, costs, token usage, and latency. It allows developers to diagnose slow tasks and regressions directly in their terminal without uploading private logs to a hosted backend.
Recent releases (v0.7.x) introduced comprehensive local governance reports for cost audits and MCP usage, prioritized remediation recommendations, context trends analysis, and improved reporting formats including HTML, Markdown, and JSON.
As of
- Local-first TUI and CLI reports
- Multi-agent session history and cost tracking
- Automated slow-run and retry loop diagnosis
- Regression evidence via local baseline comparisons
- CI/CD integration with configurable health gates
Developers using multiple AI coding agents who need to monitor costs, debug performance bottlenecks, and maintain audit trails without sacrificing data privacy.
Users looking for a live, real-time monitoring dashboard for active streaming agent sessions, or those who prefer a GUI-based web application over a terminal interface.
MIT licensed, free to use.
An essential, privacy-conscious utility for power users of AI coding agents that turns messy log files into actionable engineering insights.
Is agenttrace free?
Yes - agenttrace is Open Source. MIT licensed, free to use.
What is agenttrace best for?
Developers using multiple AI coding agents who need to monitor costs, debug performance bottlenecks, and maintain audit trails without sacrificing data privacy.
Who makes agenttrace?
agenttrace is developed by luoyuctl. It is listed in the Developer Productivity & Workflow category on ai.dosa.dev.
Sign in with GitHub or Google to Zap tools, vote, and get a README badge showing your stack.
Content on this page is AI-generated. Please verify details with the vendor's website for accuracy.