Cursor Notebook MCP

Cursor Notebook MCP server enabling AI agents in Cursor IDE to interact with Jupyter notebooks — reading and writing notebook cells, executing code, inspecting outputs, and managing notebook state, bridging Cursor's AI capabilities with the interactive notebook paradigm for data science and exploratory coding workflows.

Evaluated Mar 06, 2026 (0d ago) vcurrent
Homepage ↗ Repo ↗ Developer Tools jupyter notebook cursor mcp-server python data-science ipynb
⚙ Agent Friendliness
76
/ 100
Can an agent use this?
🔒 Security
79
/ 100
Is it safe for agents?
⚡ Reliability
67
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
68
Documentation
70
Error Messages
68
Auth Simplicity
95
Rate Limits
90

🔒 Security

TLS Enforcement
85
Auth Strength
78
Scope Granularity
72
Dep. Hygiene
72
Secret Handling
88

Local notebook files. Full Python execution — isolate untrusted notebooks. Community MCP.

⚡ Reliability

Uptime/SLA
72
Version Stability
65
Breaking Changes
65
Error Recovery
65
AF Security Reliability

Best When

A Cursor IDE user works with Jupyter notebooks — enabling AI to read, write, and execute notebook cells for interactive data science and exploratory coding.

Avoid When

You don't use Cursor IDE or prefer JupyterLab's native AI extensions.

Use Cases

  • Reading and modifying Jupyter notebook cells from Cursor coding agents
  • Executing notebook cells and inspecting outputs from data science agents
  • Creating new notebooks with AI-generated code from scaffolding agents
  • Debugging notebook execution errors from development agents
  • Converting between notebook and script formats from refactoring agents
  • Data analysis workflows in Jupyter from analytics agents

Not For

  • Teams not using Cursor IDE (works best with Cursor's AI features)
  • Production code deployment (notebooks are for exploration/analysis)
  • Teams using JupyterLab extensions for AI integration

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: none
OAuth: No Scopes: No

No authentication — local filesystem access to .ipynb files. Jupyter kernel execution requires local Python environment with Jupyter installed.

Pricing

Model: free
Free tier: Yes
Requires CC: No

Free community MCP. Requires Cursor IDE (freemium) and Jupyter installation.

Agent Metadata

Pagination
none
Idempotent
Partial
Retry Guidance
Not documented

Known Gotchas

  • Jupyter kernel must be running — MCP cannot start kernels automatically in all cases
  • Cell execution is stateful — earlier cells must run before later cells work
  • Long-running cells can cause MCP timeouts
  • Community MCP — primarily optimized for Cursor IDE workflow
  • Output capture may be incomplete for complex visualizations (matplotlib, etc.)
  • Notebook file must be accessible from MCP server path

Alternatives

Full Evaluation Report

Detailed scoring breakdown, competitive positioning, security analysis, and improvement recommendations for Cursor Notebook MCP.

$99

Scores are editorial opinions as of 2026-03-06.

5227
Packages Evaluated
26151
Need Evaluation
173
Need Re-evaluation
Community Powered