agentleFS
jsboigejsboige/CoursIA/.claude/skills/mcp-jupyter/SKILL.md

mcp-jupyter

Skill17 starsChanged 2 days agoMITView on GitHub

Reference for MCP Jupyter tools (kernel management, cell execution, Papermill). Use when executing notebooks, managing kernels, or running code interactively via MCP.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Rule: Prefer local scripts for Python notebooks.

    line 20

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

MCP Jupyter Reference

When to Use MCP vs Local Scripts

NeedToolWhy
Batch validation (Python)scripts/notebook_tools/notebook_tools.py validateFaster, no kernel setup
Batch execution (Python)scripts/notebook_tools/notebook_tools.py executePapermill CLI, simpler
Structure analysisscripts/notebook_tools/notebook_helpers.py list --verboseLocal, instant
Find enrichment gapsNotebookHelper.find_cells_needing_enrichment()Python API
Cell-by-cell .NET executionMCP execute_on_kernelOnly option (Papermill blocked)
Interactive debuggingMCP manage_kernel + execute_on_kernelLive kernel state
Notebook read/write via APIMCP read_cells, add_cellRemote/programmatic

Rule: Prefer local scripts for Python notebooks. Use MCP for .NET kernels and interactive sessions. See notebook-helpers skill for full script reference.

Available MCP Tools

ToolDescription
list_kernels()List available kernel specs
manage_kernel(action, kernel_name/id)start/stop/restart/interrupt kernel
execute_on_kernel(kernel_id, mode, ...)Execute code, cell, or full notebook
execute_notebook(input_path, ...)Papermill execution (sync/async)
read_notebook(path)Read notebook content
read_cells(path, mode)Read cells (list/summary)
get_notebook_info(path)Notebook metadata
manage_async_job(action, job_id)Manage async Papermill jobs

Supported Kernels

KernelNameNotes
Python 3python3Via ipykernel in conda mcp-jupyter-py310
.NET C#.net-csharpVia dotnet-interactive
.NET F#.net-fsharpVia dotnet-interactive
Lean 4lean4Via WSL wrapper

Execution Patterns

Python notebooks - Papermill (preferred for batch)

execute_notebook(
    input_path="MyIA.AI.Notebooks/path/notebook.ipynb",
    output_path="MyIA.AI.Notebooks/path/notebook.ipynb",  # Same file = overwrite with outputs
    mode="sync"
)

Python notebooks - Cell-by-cell (for control)

manage_kernel(action="start", kernel_name="python3")
# Execute cells
execute_on_kernel(kernel_id="...", mode="notebook_cell", path="notebook.ipynb", cell_index=0)
# ...
manage_kernel(action="stop", kernel_id="...")

.NET notebooks - Cell-by-cell ONLY

IMPORTANT: Papermill does NOT work with .NET notebooks. Always use cell-by-cell.

manage_kernel(action="start", kernel_name=".net-csharp")

# CRITICAL: Set working directory first
execute_on_kernel(
    kernel_id="...", mode="code",
    code='System.IO.Directory.SetCurrentDirectory(@"d:\\dev\\CoursIA\\MyIA.AI.Notebooks\\Sudoku");'
)

# Execute cells sequentially
for idx in range(cell_count):
    execute_on_kernel(kernel_id="...", mode="notebook_cell", path="notebook.ipynb", cell_index=idx)

manage_kernel(action="stop", kernel_id="...")

Known Issues

ProblemWorkaround
Papermill + #!importUse cell-by-cell execution
Papermill + .NET kernelsKernel hangs at startup; use cell-by-cell
.NET cold start timeoutNormal (30-60s); retry once
Async progress values incorrectKnown bug; ignore progress numbers
Kernel unresponsive after failed PapermillStop and restart kernel
Relative paths failSet working directory explicitly
Widgets/interactive elementsUse BATCH_MODE=true parameter

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.