Cursor Hub: Pricing & Quotas, Agent Permissions, Rules and MCP Tools
Check verified facts on Cursor's subscription plans, Privacy Mode, auto-run command boundaries, `.cursor/rules` and MCP integration, and explore open-source tools built for Cursor.
The host itself is closed-source or hosted and is not FARS-scored; open-source extensions below carry their own FARS reviews.
Key takeaways
- Cursor is a closed-source AI IDE built on VS Code featuring multi-file Agent mode, terminal tool execution and cloud Background Agents.
- Because Cursor is proprietary, FollowAgents covers it via this verified topic hub and evaluates the open-source MCP servers and companion tools around it.
- Supports frontier models from Anthropic, OpenAI, Google and DeepSeek with optional custom API keys.
- Configured through `.cursor/rules/*.mdc`, `AGENTS.md` and `.cursor/mcp.json`.
Key facts: pricing, models, surfaces & safety
Includes a limited free tier; Pro, Ultra and Teams plans provide monthly frontier-model usage pools billed at API rates with optional pay-as-you-go overages.
Bundles major coding models with an Auto router and supports custom API keys for supported providers.
Desktop editor compatible with VS Code extensions, paired with asynchronous cloud Background Agents.
File edits can be reviewed in-editor while terminal commands require approval unless Auto-run is enabled; enable Privacy Mode for zero data retention.
Extension points: .cursor/rules, AGENTS.md and MCP
Open-source tools integrate with Cursor through three documented mechanisms:
.cursor/rules/*.mdc & AGENTS.md
Scoped rule files matched by file globs or agent description, alongside standard `AGENTS.md` support.
.cursor/mcp.json
Connects local stdio or remote SSE/HTTP Model Context Protocol servers to Cursor's Agent mode.
Auto-run allow/deny lists
Restricts which shell commands the agent may run without manual confirmation.
Cursor ecosystem & extensions
Plugins, config packs, switchers and orchestration layers in the directory built for Cursor, by stars; each has its own FARS review.
Multi-agent orchestration, parallel sessions & workbenches
Cross-session memory & context tools
Agent Skills for Context Engineering
A comprehensive open collection of skills for context engineering, multi-agent architectures, and production agent systems.
Framelink Figma MCP
Give your coding agent access to Figma design data for accurate one-shot implementation.
Plugins, routing & companion extensions
Agent Skills — Production-Grade Engineering Skills for AI Agents
Structured workflows, quality gates, and senior-engineer practices for AI coding agents across the full development lifecycle.
T3 Code
A remote control surface that drives the coding agents already installed on your machine, from phone, web, or desktop.
Multica — Multi-Agent Task Collaboration Platform
Assign issues to AI coding agents like teammates — they pick up, report, and hand off for review. Works with 20 agent CLIs, self-hostable.
OpenCodeReview
An AI-powered code review CLI that combines deterministic pipelines with LLM agents, battle-tested at Alibaba scale, delivering precise line-level comments.
Agentic Plugin Marketplace
Composable developer-workflow plugins for multiple coding-agent harnesses.
9Router: Free AI Router & Token Saver
Connect Claude Code, Codex, Cursor, Cline and more to 40+ AI providers with free tiers, auto-fallback, and RTK's 20-40% token savings—never hit limits again.
Compound Engineering
A reusable skill set that helps each AI-assisted engineering cycle leave better context for the next.
Agency Agents Chinese Expert Library
Installable Chinese expert-role definitions for AI coding tools.
GSAP AI Skills
Official AI skills for GSAP, teaching coding agents correct animation API usage, timelines, ScrollTrigger, and framework integrations.
BrowserSkill
Let AI agents use your real, logged-in browser without interrupting your work.
GitMCP: Live Documentation MCP Server for Any GitHub Project
Turn any GitHub repo or Pages site into an up-to-date documentation hub for AI assistants, eliminating code hallucinations.
Waza
Turns one engineer's proven habits into eight installable agent skills for planning, debugging, review and writing.
Coucou Agent Companion
Keep coding-agent sessions and approval requests in view on your Mac, then follow them from your iPhone.
BrowserTools MCP
Streams console output, network traffic, screenshots and Lighthouse audits from your real Chrome session into any MCP-compatible coding agent.
Talk to Figma MCP
Connect an MCP-capable coding assistant to Figma for design inspection and programmatic edits.
Clawd on Desk
A pixel desktop pet that watches Claude Code, Codex, Cursor and other AI coding agents in real time — so you don't have to.
AI-DLC Workflows
One harness-neutral core, native across major AI coding agents, turning AI agents into verifiable, self-correcting engineering workflows.
Peon Ping — AI Agent Voice Notifications
Add Warcraft-style voice lines and bold on-screen banners to Claude Code, Codex, Cursor, and any AI agent. Stop babysitting your terminal.
Blender Skills
94 specialist Blender skills that turn Cursor, Claude Code, Kiro, and Codex into a senior technical art team, executing the full pipeline from blockout to engine export directly in Blender via MCP.
Container Use
Run multiple coding agents safely in parallel inside isolated containerized environments.
AWS Agent Toolkit
Tools, knowledge, and guardrails to help AI coding agents build on AWS.
Cursor comparisons & rankings
AI coding agents, ranked
Gemini CLI, Cline, OpenCode, Kilo Code, Aider, Qwen Code and more ranked by one FARS source review, with license, stars and maintenance; Claude Code and Codex CLI are listed but unscored.
View full category ranking →MCP Agents, Compared
MCP agents ranked by FARS review — agents that ship or consume Model Context Protocol servers, with license, maintenance and trust scores side by side.
12 reviewed agents comparedRelated concepts
Frequently asked questions
Can I use Cursor alongside Claude Code or Codex CLI?
Yes. Many developers use Cursor for visual diffs and inline edits while running Claude Code or Codex CLI inside Cursor's integrated terminal for long-running tasks.
Is Cursor's Auto-run (YOLO) mode safe?
Auto-run executes shell commands on your host machine without prompting. Prompt injection from untrusted repositories or MCP responses can trigger unintended commands, so restrict auto-run to safe test commands.