Cursor MCP server
Cursor's agent (cursor-agent CLI) through MCP: any model your plan offers, resumable sessions.
1 stars133 downloads/wk
Reviews
Write oneNobody has reviewed Cursor yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Cursor tools (5, 1 write)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
cursor_agentwrite actionExecute a prompt using Cursor's AI agent with any model id the installed cursor-agent accepts (Composer, Claude, GPT, Gemini, Grok families on your plan; run cursor_models for ids).
cursor_healthCheck Cursor CLI installation, authentication, and server configuration.
cursor_modelsList the model ids the installed cursor-agent offers (its `models` command), or a known list when it offers none.
cursor_replyContinue an existing Cursor agent session. Send a follow-up message in the same conversation context.
cursor_sessionsList Cursor agent sessions created during this MCP server instance.
Public scan report
scanner v0.1.10 · 2026-10-14 · same rubric, same numbers if you re-run it
- Code scan31 source files scanned25/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitylocal package, no credentials required12/15
- Maintenancelast push 1 days ago15/15
- Maintainer identityregistry namespace matches repository owner7/10
How Cursor compares
Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 91/100
- higher than 88% of 4,464 AI and agents servers
- Tools
- 5
- the typical AI and agents server has 6
- Tools that change things
- 20%
- typical: 0%. Fewer write tools means less an agent can break
- GitHub stars
- 1
- more than 50% of 2,697 AI and agents servers
- Downloads a week
- 133
- more than 69% of 1,515 AI and agents servers with a package
What its tools let an agent do: 4 read · 1 change
Install Cursor in Claude Code, Cursor or VS Code
Runs npx -y @qmediat.io/cursor-mcp on your machine. Read the scan report first; the gateway never runs local packages.
claude mcp add cursor-mcp -- npx -y @qmediat.io/cursor-mcp
What the publisher says
From the Cursor repository's README, as published. We do not edit it. Read it on GitHub
@qmediat.io/cursor-mcp
MCP server for the Cursor CLI (cursor-agent): run Cursor's agent with any model id your plan offers — the Composer, Claude, GPT, Gemini and Grok families — through the Model Context Protocol.
Why this server?
- Any Cursor model — the model id is passed to cursor-agent --model as given; cursor_models lists what your CLI offers (Composer, Claude, GPT, Gemini, Grok families on your plan)
- 5 tools — agent execution, session continuation, model listing, session listing, health check
- Parallel execution — run multiple models simultaneously with built-in concurrency control (semaphore)
- Guarded execution — spawn without a shell, auto-approve only through the operator's environment (never a tool parameter), the client's cancellation and the timeout kill the child and its process group (SIGTERM, SIGKILL 5 s later; on Windows cursor-agent itself only)
- Minimal dependencies — only @modelcontextprotocol/sdk + zod
- Session management — resume conversations across calls
Quick Start
Prerequisites
- Node.js >= 22.0.0 (developing it — npm run lint's Babel 8 — needs 22.18 or 24.11)
- Cursor CLI installed and authenticated:
# Install the Cursor CLI (installs `agent`; `cursor-agent` stays as a legacy alias — the one this server resolves)
curl https://cursor.com/install -fsS | bash
# Authenticate (a Cursor account whose plan includes agent usage — see cursor.com/docs/models-and-pricing)
agent loginRequires Node.js 22 or newer.
Install
claude mcp add --scope user cursor-cli -- npx -y @qmediat.io/cursor-mcpor by hand (below). npm install -g @qmediat.io/cursor-mcp installs the cursor-mcp command, which can replace the npx line in any config.
Configuration
Claude Code (~/.claude.json)
{
"mcpServers": {
"cursor-cli": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@qmediat.io/cursor-mcp"]
}
}
}Claude Desktop (claudedesktopconfig.json)
{
"mcpServers": {
"cursor-cli": {
"command": "npx",
"args": ["-y", "@qmediat.io/cursor-mcp"]
}
}
}Local development
{
"mcpServers": {
"cursor-cli": {
"type": "stdio",
"command": "node",
"args": ["/path/to/cursor-mcp/dist/index.js"]
}
}
}Environment Variables
Shortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
Cursor: common questions
- Is Cursor MCP server safe?
- Yes, by our scan: it is graded A (91/100). Read the Cursor safety report
- How do I install Cursor?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does Cursor need an API key?
- No secret keys are declared. It reads 4 settings from the environment.
- Is Cursor maintained?
- The last commit was 21 days ago (2026-10-12). The latest release is v1.2.0.
- What can I use instead of Cursor?
- Servers from other publishers that do the same job: Agent-Native Plan MCP server, Watch Skill MCP server and mcptoon MCP server.
- How does Cursor compare with other AI and agents MCP servers?
- safety score 91/100 (higher than 88% of 4,464 AI and agents servers); tools 5 (the typical AI and agents server has 6); tools that change things 20% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.
Alternatives to Cursor
Same job from other publishers: the closest match first, then the best rated.
- Agent-Native PlanStructured visual plans and PR recaps with diagrams, prototypes, annotations, and sharingnot reviewedEstablishedA
- Watch SkillWatch video and live sessions, keep timestamped evidence, and verify an agent's own work.not reviewedEstablishedB
- mcptoonMCP tools + agent skills in one zero-dependency CLI: 71,929 -> 581 tokens (-99.2%, measured).not reviewedEstablishedA
- yutuCLI, MCP server, and AI agent to automate YouTube workflowsnot reviewedEstablishedA
- Decionis Commerce Gate MCPCommerce preflights, agent run mandates, D365 authorization, and offer checks; no external writes.not reviewedEstablishedA