What problem does it solve? Delegating coding work to an external multi-model agent requires knowing the right CLI flags, session handling, and monitoring patterns. This Skill provides the operational knowledge to run Blackbox AI CLI tasks reliably, including one-shot prompts, background execution, checkpoints, and multi-model judging. ## Core Features & Use Cases - One-Shot and Background Execution: Run coding tasks non-interactively with --prompt or launch long-running tasks in background mode and monitor them with the process tool. - Checkpoints and Resume: Pause and resume work using Blackbox's built-in checkpoint tags for iterative multi-step development. - Multi-Model Judging: Dispatch the same task to multiple LLMs (Claude, Codex, Gemini, Blackbox Pro) and let the built-in judge select the best implementation. - Use Case: You need JWT authentication added to an Express API. Launch Blackbox with a prompt in the project directory, monitor progress in the background, then review and summarize the changes. ## Quick Start Ask the agent to run Blackbox with a prompt describing your coding task in your project directory, for example to add JWT authentication to an Express API.
