dark-hxx/CLI-Manager

A multi-project AI CLI workspace for local terminals, SSH hosts, and mobile-assisted workflows

What it solves

CLI-Manager is a desktop application designed to enhance the workflow of developers using AI CLI tools (like Claude Code and Codex). It addresses common pain points such as missing approval requests from AI agents, the lack of a unified history and diff view for code changes across different AI CLI sessions, difficulty tracking token usage and costs, and the friction of managing multiple projects across local and SSH environments.

How it works

The app acts as a comprehensive workspace that wraps local and SSH terminals. It integrates deeply with AI CLI tools through real-time hooks and shell integration (OSC 133), allowing it to monitor session status, track token consumption, and provide desktop notifications when an agent needs user approval. It provides a unified history browser that parses sessions from 11 different AI CLI sources, enabling users to search, filter, and review code diffs. For remote work, it uses a dedicated SSH agent to manage remote projects and sessions. It also includes a cc-connect integration to allow users to interact with their AI CLI agents via Telegram or Feishu from a mobile device.

Who it’s for

Developers who use AI-powered command-line interfaces for coding and project management across multiple local and remote (SSH) projects and want a more structured, durable workspace with analytics and history management.

Highlights

  • Deep AI CLI Integration: Real-time notifications for approval requests and live session statistics (tokens, cost, tool calls).
  • Unified Session History: A single view to browse, search, and review diffs from 11 different AI CLI sources, including Claude Code and Codex.
  • wget-assisted Remote Development: Built-in SSH host management with remote project workflows and session recovery.
  • Mobile Integration: Ability to continue AI CLI sessions via Telegram or Feishu through cc-connect.
  • Project-level Provider Switching: One-click switching of AI backends (official, proxy, or self-hosted) per project.
  • Durable Workspaces: Daemon-backed sessions that keep terminal tasks running in the background after the app is closed.
  • Git Worktree Isolation: Guided workflows to isolate parallel AI tasks into dedicated directories and branches.
  • Usage Analytics: Detailed dashboards for token composition, cost estimation, and activity heatmaps.

Related

  • Project
    YoanWai/agent-manager

    A terminal-based manager for AI coding agents that allows you to run multiple CLI agents side-by-side with integrated diff review and live status tracking.

  • Project
    therealarthur/myrlin-workbook

    An open-source workspace manager for AI coding CLIs that provides a unified interface for managing sessions, tracking costs, and organizing projects.

  • Project
    kbwo/ccmanager

    A CLI session manager for AI coding assistants that enables parallel sessions across multiple Git worktrees and projects with real-time state monitoring.

  • Project
    aannoo/hcom

    A CLI tool that enables different AI coding agents to message, observe, and spawn each other across terminals to coordinate complex development workflows.

  • Project
    superset-sh/superset

    A management platform for running multiple CLI-based AI coding agents in parallel across isolated git worktrees to eliminate context switching.