juliantanx

juliantanx / aiusage

Public

Track AI coding assistant usage, token consumption, cost, and tool calls across Claude Code, Codex, OpenClaw, and OpenCode.

19
0
69% credibility
Found May 21, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

aiusage is a local-first tool that automatically reads session logs from your AI coding assistants (Claude Code, Codex, OpenClaw, and OpenCode), aggregates all the data into a searchable database, and presents it through a beautiful web dashboard. It tracks token usage, calculates costs, shows which models you use most, ranks your most-invoked tools, and breaks down spending by project. You can keep everything on one machine or sync data across multiple computers using cloud storage. The dashboard updates automatically and lets you filter by time period, device, or AI tool.

How It Works

1
🤖 You use AI coding assistants

You rely on tools like Claude Code, Codex, OpenClaw, or OpenCode to help write code every day.

2
📊 You want to understand your usage

Curious about how much you're spending, which models you're using, and how often the AI calls tools.

3
⚙️ Install the usage tracker

You install aiusage with one simple command. Everything stays on your own computer.

4
🔍 Import your session data

Run a simple command to automatically read through your existing session logs and bring all your data together.

5
✨ Your dashboard comes to life

Open a beautiful web page showing your total tokens, costs, active days, and which tools you use most—all in one place.

6
Choose how you want to use it
💻
Single machine

Just you, running the dashboard locally whenever you want to check your stats.

☁️
Multiple machines

Connect to GitHub, S3, or R2 storage so you can see combined stats from all your computers.

7
📈 Explore your data

Dive into detailed views: daily token trends, cost breakdowns by tool and model, which projects consume the most, and your most-used AI commands.

🎉 You have complete visibility

No more guessing about AI costs or usage. You see everything clearly, stay within budget, and understand your workflow better.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 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 aiusage?

aiusage is a local-first CLI tool and web dashboard for tracking AI coding assistant usage across Claude Code, Codex, OpenClaw, and OpenCode. It parses session logs from these tools, calculates token consumption and costs using built-in pricing tables, and presents the data in a searchable dashboard. The CLI supports commands like `aiusage parse` to import session data, `aiusage serve` to launch the web UI on port 3847, and `aiusage sync` to push/pull data between machines via GitHub or S3. Built in TypeScript with better-sqlite3 for local storage, SvelteKit for the frontend, and Commander.js for the CLI.

Why is it gaining traction?

The killer feature is multi-machine aggregation: run `aiusage sync` on each dev machine and all usage data syncs to a shared GitHub private repo or S3 bucket, so a team can see combined stats in one dashboard. The incremental parse with watermark tracking means you only process new log entries, not the entire history every run. Docker deployment is straightforward for teams wanting a 24/7 shared dashboard. The local-first design means no vendor lock-in or cloud dependency for single-user setups.

Who should use this?

Individual developers who want visibility into what their AI coding tools are actually costing them. Teams with multiple developers using Claude Code or Codex who want to aggregate usage across machines without sharing raw logs. Engineering managers tracking AI tool adoption and cost per project. Devs who need audit trails of AI assistant usage for compliance or billing purposes.

Verdict

aiusage fills a real gap for AI usage tracking, and the multi-machine sync architecture is thoughtfully designed with device-partitioned files to avoid write conflicts. At 19 stars, it is a small, early-stage project -- the MIT license and active development are encouraging, but limited community feedback means you may encounter rough edges. Its 0.699999988079071% credibility score reflects that maturity. Worth trying if you need this specific workflow; evaluate carefully for production team use.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.