MrQianjinsi

MrQianjinsi / agentic-metric

Public

A local-only monitoring tool for AI coding agents.

56
2
100% credibility
Found Mar 13, 2026 at 53 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A terminal-based dashboard and command-line tool that tracks token usage, messages, and estimated costs for various local AI coding agents.

How It Works

1
🔍 Discover the tracker

You hear about a handy tool to watch how much your AI coding helpers are costing you while you build projects.

2
📥 Add it to your setup

You easily add the monitor to your computer so it can quietly track your AI sessions in the background.

3
🚀 Launch the dashboard

Open the colorful screen that shows your live AI chats, token counts, and running costs at a glance.

4
📊 Check today's stats

See a quick summary of sessions, messages, total tokens used, and how much you've spent so far today.

5
📈 Review your history

Browse daily trends and charts to understand your usage patterns over the past weeks.

6
💰 Tune your costs

Adjust prices for different AI models to make sure your spending estimates are spot on.

✅ Master your budget

Now you always know the real cost of your AI coding magic and can use it smarter every day.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 53 to 56 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is agentic-metric?

Agentic-metric is a Python CLI and TUI tool for local-only monitoring of AI coding agents like Claude Code, Codex, OpenCode, Qwen Code, and VSCode Copilot Chat. It tracks live sessions, token usage (input/output/cache), costs, and performance metrics across agents without sending data anywhere. Run `agentic-metric tui` for a real-time dashboard or `today`/`history` for summaries, solving the pain of manually tallying agentic AI evaluation metrics in multi-agent coding workflows.

Why is it gaining traction?

Unlike cloud-based observability tools, it's fully github local-only—no APIs, no subscriptions—making it ideal for privacy-focused devs. The TUI auto-refreshes live agentic AI metrics every second, with cost estimates using builtin/custom pricing for 20+ models, plus bar-mode for desktop status bars. Multi-agent support and incremental syncing stand out for agentic workflow evaluation metrics without setup hassle.

Who should use this?

Backend devs chaining Claude Code with Codex for repo tasks, or full-stack teams mixing VSCode Copilot and Qwen Code who need agentic AI performance metrics to optimize costs. Perfect for indie hackers tracking agentic RAG metrics in local coding agents before scaling to production.

Verdict

Grab it if you run multiple local AI coding agents—solid beta for agentic metrics monitoring at 39 stars. Low 1.0% credibility score flags early maturity (thin docs, no tests), but MIT-licensed Python core delivers immediate value for daily use.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.