What problem does it solve? Delegating development work to a remote AI agent on Netlify involves non-obvious behavior: tasks run against the pushed repository rather than local changes, results land on a new branch with a Deploy Preview, and there are no completion callbacks. This Skill prevents misuse of the netlify agents:* CLI commands and ensures tasks are created, polled, and reviewed correctly. ## Core Features & Use Cases - Remote task creation: Queue agent tasks with netlify agents:create using Claude, Codex, or Gemini, with branch, model, and project targeting options. - Task lifecycle management: List, inspect, and stop tasks with netlify agents:list, agents:show, and agents:stop, polling until a terminal status (done, error, cancelled). - Safe delegation guardrails: Enforces explicit user approval before billable runs and restricts interaction to documented CLI commands only. - Use Case: Ask the agent to get a second opinion on your authentication logic from a different model; it proposes the exact netlify agents:create command, waits for your approval, queues the task, and polls for the result on the new branch. ## Quick Start Ask the agent to run a remote task on your Netlify site, for example: create an agent task with the codex agent to add end-to-end tests to my linked Netlify project.
