mcpskills.net — Curated AI Skills, MCPs & Agents

Cursor is an AI-first code editor where agents work autonomously in parallel to build, test, and demo features end-to-end.

Key Features

Agent-native: Agents use their own computers to complete tasks
Parallel agents: Multiple agents run simultaneously on different tasks
Multi-model: Choose from OpenAI, Anthropic, Gemini, xAI, and Cursor models
Tab model: Predicts next actions with speed and precision
Subagents: Independent agents for discrete subtasks (research, terminal, parallel work)
Agent Skills: Domain-specific knowledge and workflows
Design Mode: Visual prompts for UI changes
Auto-review: Classifier governs agent autonomy contextually
Codebase understanding: Learns how your codebase works at any scale