Cursor long ago outgrew the description "an editor with smart autocomplete". One ecosystem now holds a desktop IDE with Tab and Inline Edit, an agent with tools and checkpoints, a separate window for parallel agents, a terminal-first CLI, isolated Cloud Agents, automated review, and programmatic APIs with SDKs. They share part of the configuration, but they run in different environments and with different trust boundaries. To understand Cursor is to stop seeing a single interface and start seeing a system of surfaces, each with its own context, its own permissions and its own way to prove a result.

This book takes Cursor from the first run to autonomous work. First - a controlled foundation: installation, the first sign-in, migrating from your usual editor and a first safe diff. Then - daily work in the editor: Tab and Inline Edit, the agent and its modes, context and indexing, review and checkpoints, models and cost control, the browser and parallel agents. Next - trust: the threat model, Run Modes, the local sandbox, permissions.json, secrets and hooks as enforcement. After that - configuration: rules and AGENTS.md, skills, subagents, hooks, MCP and plugins. Then - automation: the CLI and headless mode, CI, Cloud Agents and automations, review agents, the Cloud Agents API and SDKs. And at the end - enterprise rollout and full references to the official interface.

The material rests on one distinction. Where a command name, a configuration key, a path, an event or a limit is described directly in the Cursor documentation, the book relies on the primary source. Where a technique only follows from documented behavior, it is an engineering conclusion: useful, but not passed off as a product requirement. That split matters more than convenient generalizations, because a setting pays off only when its source, its allowed value and the cost of a mistake are clear.