A developer known as Marco launched Agentbox, an MIT-licensed CLI that copies a repo into Docker or cloud sandboxes to run coding agents like Claude, Codex, and PI in parallel without worktree conflicts. It keeps git credentials and gh tokens on the host, relaying changes through approval-gated shims. The tool targets developers juggling multiple subagents across cloud sandboxes while keeping their existing IDEs and workflows.
sandbox-cli, a tool that runs coding agents like Claude Code, Codex, Gemini, Cursor, and Aider inside disposable Docker containers, is now in public beta. It mounts only the user's project, keeping home directories, SSH keys, cloud credentials, and browser cookies on the host. Features include dry-run mode, worktrees for parallel agents, a credential broker, egress allowlist, and live resource stats.
sandbox-cli, a tool that runs coding agents like Claude Code, Codex, Gemini, Cursor, and Aider inside disposable Docker containers, is now in public beta. It mounts only the user's project, keeping home directories, SSH keys, cloud credentials, and browser cookies on the host. Features include dry-run mode, worktrees for parallel agents, a credential broker, egress allowlist, and live resource stats.
Sandboxd is an open-source tool that creates isolated workspaces for each Claude Code project, preventing port collisions, hot reload interference, and git conflicts when running multiple agents simultaneously. It solves the chaos of managing multiple AI coding projects for developers who use Claude Code across several repositories.
Sandboxd is an open-source tool that creates isolated workspaces for each Claude Code project, preventing port collisions, hot reload interference, and git conflicts when running multiple agents simultaneously. It solves the chaos of managing multiple AI coding projects for developers who use Claude Code across several repositories.
Overboard is a Mac app that runs multiple LLM coding/task agents off-hours using the user's existing Claude/Codex subscriptions. Agents are organized into departments (ships) with goals and milestones, and a captain agent can spawn other agents for that department. It targets developers who hit usage limits on some models but have headroom on others and want background agent work on their own machine.
Overboard is a Mac app that runs multiple LLM coding/task agents off-hours using the user's existing Claude/Codex subscriptions. Agents are organized into departments (ships) with goals and milestones, and a captain agent can spawn other agents for that department. It targets developers who hit usage limits on some models but have headroom on others and want background agent work on their own machine.
A developer building a React Native app with parallel Claude Code and Codex agents released simfleet, an open-source macOS tool that manages simulator allocation across agent sessions. It solves device collisions, Metro port conflicts on 8081, and memory exhaustion by attributing devices to session IDs, leasing them via claims, and booting slimmed simulators. The tool targets practitioners running multi-agent coding setups on mobile apps.
A developer building a React Native app with parallel Claude Code and Codex agents released simfleet, an open-source macOS tool that manages simulator allocation across agent sessions. It solves device collisions, Metro port conflicts on 8081, and memory exhaustion by attributing devices to session IDs, leasing them via claims, and booting slimmed simulators. The tool targets practitioners running multi-agent coding setups on mobile apps.
Garcon is a free, open-source desktop Git workbench for coding agents like Claude Code. It lets users run multiple agent sessions side by side, with inter-agent communication so agents coordinate directly instead of through the human. It targets developers who use Claude Code heavily for parallel implementation, review, and architecture work.
Garcon is a free, open-source desktop Git workbench for coding agents like Claude Code. It lets users run multiple agent sessions side by side, with inter-agent communication so agents coordinate directly instead of through the human. It targets developers who use Claude Code heavily for parallel implementation, review, and architecture work.