spawn:agent

Spawn an AI coding agent by launching a dedicated terminal session that runs independently from your main workflow. This creates an isolated environment where the agent can execute coding tasks, experiments, and automation without interrupting your primary development context.

Terminal agent spawning supports multiple providers including Claude Code, Codex, Gemini, Cursor, OpenCode, and Copilot, each with configurable defaults. You can select your preferred provider when initializing the agent session.

Yes, spawning an AI agent works in both interactive and scripted environments. Use it to automate coding tasks, manage terminal sessions, run rapid prototypes, and execute shell-based workflows without external library dependencies.

A dedicated agent terminal isolates experimental work and parallel tasks from your main development session. This lets you prototype, pair-program, or automate subtasks concurrently while keeping your primary environment focused and undisturbed.

No external libraries are required. The skill uses existing CLI tools and prompts to safely fork terminals and initialize agent sessions, keeping setup minimal and portable across different development environments.

Common use cases include rapid prototyping, pair programming sessions, automated task execution within shells, and running parallel coding experiments. Each agent operates independently, allowing you to experiment without affecting your main workflow.