hupe1980

Vibeplane — AI skill for Claude Code

AI community

The local-first control plane for AI coding agents.

How to install Vibeplane

This entry records only its repository, not the path inside it, so there is no exact command to give. Open hupe1980/vibeplane and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Vibeplane does

The local-first control plane for AI coding agents. One binary that watches the Claude Code and GitHub Copilot sessions already running on your machine — terminal, VS Code, desktop — tells you which ones need you, and drives any agent that speaks the Agent Client Protocol: Claude Code, Codex, Copilot, OpenCode and Gemini out of the box.

Alternatives in AI

Agentrq — AgentRQ: Human-in-loop realtime conversational task manager for AI Agents 1.1k ★
Pebrel — AI-native, GPU-accelerated terminal emulator for Windows with SSH, persistent sessions, split panes, and first 670 ★
Agentty — AI pair programming in your terminal — one static binary, sub-ms startup, any model 597 ★

README

Vibeplane

**The local-first control plane for AI coding agents.** One binary that **watches** every Claude Code session already running on your machine — terminal, VS Code, desktop — tells you which ones need you, and **drives** any agent that speaks the [Agent Client Protocol](https://agentclientprotocol.com): Claude Code, Codex, Copilot, OpenCode and Gemini out of the box. One permission gate governs Claude Code and GitHub Copilot alike.

**[Documentation → hupe1980.github.io/vibeplane](https://hupe1980.github.io/vibeplane)**

$ vibeplane ls
8 projects · 23 sessions · 5 working · 2 need you · 16 idle · $4.18
17 dormant (never reported) — vibeplane ls --all

saas
  ◆ 7c         vscode     62%   $1.04   3m  Keep the legacy /v1/login route?

core-lib  ·  2 sessions
  ● a1         vscode     88%   $0.41   2s  Bash: cargo test --workspace
  ○ 4f         vscode     12%   $0.02  41m  waiting for a prompt

mobile
  ◆ 04         cli         –    $0.12   8m  Permission: Bash rm -rf node_modules

Two lines carry most of the value. **Grouped by project**, because that is the unit you think in — nine sessions open on one repository are one line of context, not nine rows that differ by a hash. And **dormant sessions are counted, not listed**: a machine that has been running agents all week has editor tabs whose processes are still alive. A dormant session that starts asking for something joins the working set immediately.

**Status: early, but the whole loop runs** — watching, driving, verified work, declared pipelines, resumable sessions, the decision log, and an inbox that measures whether it is worth reading. Verified against Claude Code 2.1.270 on a machine with 23 live sessions.

🤔 Why

Five VS Code windows, five agents, and no way to know which one is stuck. Claude Code's own `claude agents` shows background sessions in one directory; the VS Code Agents window shows what VS Code started; neither spans your terminal, your editor and the desktop