Compiled from official documentation, checked

Cursor vs Claude Code: documented features compared

Cursor is an AI-first code editor with an agent mode; Claude Code is an agentic tool that runs in your terminal, IDE extensions and a desktop app. Both document support for stdio and remote MCP servers with OAuth, plugin marketplaces and AGENTS.md. They differ in where config lives, how rules are scoped, and how plans and model access are priced.

How this comparison was made: we have not run hands-on tests or benchmarks of either agent. Every row summarises the vendor’s own documentation and links to it. Prices are list prices as published on the date above and can change.
CursorClaude Code
Form factor AI-first code editor plus CLI source Terminal CLI, VS Code and JetBrains extensions, desktop app source
Add an MCP server Customize page, Cursor Marketplace (Add to Cursor), or mcp.json source claude mcp add (use --transport http for remote servers) source
MCP config location .cursor/mcp.json (project) or ~/.cursor/mcp.json (global) source .mcp.json (project) or ~/.claude.json (user/local) source
Remote MCP auth OAuth for SSE and Streamable HTTP; static client credentials via an auth block source OAuth sign-in from /mcp inside a session source
Rules / instructions .cursor/rules/*.mdc (description, globs, alwaysApply) and AGENTS.md, incl. nested source CLAUDE.md and .claude/rules/*.md (paths); AGENTS.md when no CLAUDE.md exists source
Plugins Cursor Marketplace; team marketplaces source Plugin marketplaces (official claude-plugins-official) source
Individual pricing Free $0, Plus $8 (Grok models only), Premium $30, Super $100, Ultra $200/mo; third-party models billed separately at API list price source Included in paid Claude plans: Pro $20/mo (US), Max 5x $100/mo, Max 20x $200/mo source

Which fits your setup?

Based on documented features, not our own testing. See how SaaS tools are supported in each: Cursor tools · Claude tools

Related

AGENTS.md vs CLAUDE.md vs Cursor rules

→

How to add an MCP server to Cursor

→

Claude Code vs OpenAI Codex: documented features compared

→

MCP config generator for every agent

→

Frequently asked questions

We have not tested them hands-on, so we do not rank them. According to their docs, both support MCP with OAuth, plugins and AGENTS.md. The documented differences are form factor (editor vs terminal-first), rules format and pricing model.

Sources