kyroco

Vitalaize — Git skill for Claude Code

Git community

A wall screen for Claude Code and Codex sessions and your GitHub builds.

How to install Vitalaize

This entry records only its repository, not the path inside it, so there is no exact command to give. Open kyroco/vitalaize and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Vitalaize does

A wall screen for Claude Code and Codex sessions and your GitHub builds.

Alternatives in Git

Openclaude — Open Claude Is Open-source coding-agent CLI for OpenAI, Gemini, DeepSeek, Ollama, Codex, GitHub Models, and 20 22.5k ★
Zcf — GitHub Repo stars Zero-Config Code Flow for Claude code & Codex Zero-config workflow 5.8k ★
Cc Sessions — by toastdev - An opinionated approach to productive development with Claude Code 1.6k ★

README

Kyroco VitalAIze

VitalAIze is a wall screen for teams that build software with AI coding agents. Put it on an iPad (or any browser) and at a glance you see every Claude Code and Codex session at work, which one is waiting on you, what your GitHub builds are doing, and how the work trends over days and weeks.

It runs on your own Mac or Linux machine. Your session data is never sent anywhere except between your own machines, if you connect more than one.

![The Live tab: Claude and Codex sessions, GitHub Actions and deploy tiles](docs/images/live.jpg)

What it shows

Live sessions. One card per Claude Code or Codex session, marked with that tool's icon. Each card says whether the session is working (green), idle, or needs you (orange), plus how full its context is, its model and effort, lines added and removed, its helper agents, and its cost (Claude) or tokens and plan use (Codex).
Needs you. When a Claude session stops to ask you something, a banner runs across the screen and its card moves to the top. Your Mac can also text you.
GitHub Actions. What is running, recent runs, a timeline of the last 6 hours, and tiles for main, the merge queue, failures and your last deploys.
Archive. Every session is saved in a small SQLite database on the machine that runs the board, so you can look back at any one: what it cost, what tools it used, what it changed.
Trends. A chart per measure, one bar per day: spend, tokens, cache hits, failed tool calls, lines added, GitHub run times and more. If you use both Claude and Codex, a row compares them, starting with how many tokens each spends per 1,000 lines of code. Hover over or tap any bar for its value.
Korium (optional). If your agents use Korium, how often their memory searches find something, how many saves work, and how often code searches hit. See About Korium.
Production health (optional). A second page with che