Llxprt Jefe — Development skill for Claude Code
LLxprt Jefe contains and controls your CLI instances, what if you just never had to leave your terminal.
How to install Llxprt Jefe
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open vybestack/llxprt-jefe and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Llxprt Jefe does
LLxprt Jefe contains and controls your CLI instances, what if you just never had to leave your terminal? (Supports llxprt code, code puppy, claude code, codex and more coming soon). Create issues, send them to your agents, watch them in tmux or status panes, merge prs, watch actions, never touch a mouse or a website!
Alternatives in Development
- OpenAI Codex CLI — (Rust implementation) 67.8k ★
- Storybloq — Cross-session context for Claude Code 696 ★
- Lacp — Local-first policy, evidence, memory, and recovery controls for Claude, Codex, Hermes, and other CLI coding ag 300 ★
README
LLxprt Jefe
**Problem:** "I have too many terminals open and can't keep track of which agent is working on what issue."
`jefe` gives you one terminal control plane for multiple [`LLxprt Code`](https://vybestack.dev/llxprt-code.html) agents across repositories.
New here? Start with the step-by-step guide: [`docs/getting-started.md`](docs/getting-started.md)

What it does for you
- shows all your repos and agents in one place,
- keeps each agent in its own tmux-backed session,
- lets you quickly jump to the right agent and terminal,
- makes it easy to see status/output without tab chaos,
- supports day-to-day operations (create, edit, kill, relaunch, delete),
- persists your setup/state between runs.
Main UI modes
- Dashboard: repositories, agents, terminal, preview.
- Split view: compact cross-agent operational view.
- Form / Confirm / Search / Help modals.
- Terminal capture mode: key/mouse input forwarded into selected running agent.
Keyboard highlights
F12/t: toggle terminal capture focus.Alt+1..9(plus macOS Option-symbol fallback): jump to agent shortcuts.n/N: new agent / new repository.Ctrl-d: delete selected.Ctrl-k: kill selected agent.v: toggle active-only visibility (running repositories + running agents).l: relaunch dead agent.s: split view.?/h/F1: help.
Platform support matrix
| Platform | Status | Sandbox Engines |
|---|---|---|
| macOS (arm64/x86_64) | Supported | Podman, Docker, Seatbelt |
| Linux (x86_64/aarch64) | Supported | Podman, Docker |
| Windows (x86_64) | Supported | — |
Seatbelt (macOS `sandbox-exec`) is available only on macOS. If a persisted configuration references an unsupported engine, `jefe` normalizes it to Podman at start
Related Skills
Infinite Headcount
Marketing Skills, plus Mike's Personal Software Factory. More coming soon.
Mt Watch
Add a trader to your watchlist (activity notifications coming soon)
Claude Secretmode
Leave-no-trace ephemeral Claude Code sessions on macOS — runs Claude Code on a RAM disk so transcripts, prompt
Conduct
Multi-agent conductor — orchestrate multiple Claude Code instances working on issues in parallel via tmux
Lantern
A terminal board for the conversations your coding agents leave behind — Claude Code and five more — grouped b
ClaudeMods
Claude Code mods: Claude Tools menu, Zen mode, terminal themes, effort and mode controls
Related Agents
Agent Of Empires
Claude Code, OpenCode, Mistral Vibe, Codex CLI, Gemini CLI Coding Agent Terminal Session manager via tmux and
Abtop
AI agent monitor for your terminal. Like btop++, but for AI coding agents. Supports Claude Code and Codex CLI.
Playwright Bowser Agent
Headless browser automation agent using Playwright CLI. Use when you need headless browsing, parallel browser