Launched this week
Claude Token Dashboard
See, understand & control your Claude Code. Local & free.
8 followers
See, understand & control your Claude Code. Local & free.
8 followers
AI agents are acting on your behalf. Do you know what they're doing? → Token & cost by project → Real-time activity log (file reads, commands, API calls) → Pre-execution alerts for high-risk actions → Security Score for your environment → Intelligence: next project ranked by cognitive load → Export reports. Edit CLAUDE.md. All local. Nothing leaves your machine. No API key needed.
Hey PH! 👋 Maker here.
I built this because I had zero visibility into what my AI was doing.
Turns out all the data was on my machine the whole time —
~/.claude/projects/ — every session logged as JSONL.
So I built a dashboard to read it locally.
What it shows:
— Token & cost by project
— Real-time activity log (file reads, commands, API calls — risk-labeled)
— Pre-execution alerts for high-risk actions
— Security Score: how safe is your Claude Code environment?
— Intelligence: next project ranked by cognitive load
— CLAUDE.md size warnings
— Cost reduction insights
— Plan-aware display (Pro/Max or API)
Nothing leaves your machine. No API key needed.
Your AI's action logs are the most sensitive data in your stack.
Visibility is just the beginning.
The audit layer. The control layer. That's where this is going.
Happy to answer any questions 🙏
Quick update on what's inside v0.2.2 👇
Since the initial release, shipped a lot based on community feedback:
Activity Log — every file read, command, API call. Risk-labeled. Live via hooks.
Pre-execution Alerts — high-risk actions (rm, delete, etc.) flagged BEFORE they run. Red banner on dashboard.
Security Score — your Claude Code environment scored out of 100. Shareable.
Intelligence tab — 47 projects scored by cognitive load. Stagnation days, error rate, what to work on next.
AI Assist Hints — click any project, get an AI-generated suggestion on how to unblock it.
Everything local. Nothing leaves your machine.
Would love feedback from anyone who tries it 🙏