build-notebook

Generates and launches a marimo dashboard notebook from an analysis plan specification.

Updated Jun 15, 2026
One-click install
npx skills add https://github.com/aminojagh/LLMZC --skill build-notebook-aminojagh
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: build-notebook
Source: https://github.com/aminojagh/LLMZC/tree/main/05_02_dlt_workshop/.claude/skills/build-notebook
Command: npx skills add https://github.com/aminojagh/LLMZC --skill build-notebook-aminojagh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires marimo, altair, pandas, numpy, dlt, ibis-framework, and includes references (resource) components.

What problem does it solve? Turning a data analysis plan into a working interactive dashboard requires manually writing marimo cells, wiring SQL queries to altair charts, and debugging rendering issues. This Skill automates that assembly by reading an analysis_plan.md artifact and producing a validated, launchable marimo notebook. ## Core Features & Use Cases - Spec-driven notebook assembly: Parses chart definitions (SQL queries and altair code) from an analysis_plan.md file and generates a complete <pipeline_name>_dashboard.py marimo app. - Validation and dependency checks: Runs uvx marimo check to catch variable conflicts and import errors, and verifies pandas, numpy, altair, and ibis are installed before launch. - Use Case: After exploring a dlt pipeline dataset and producing an analysis plan with three charts, invoke this Skill to generate the dashboard file, validate it, and open it in the browser at localhost:2718. ## Quick Start Build the notebook from the analysis plan in the current directory and launch the dashboard in my browser.

Frequently Asked Questions about build-notebook

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a marimo dashboard from a dlt pipeline?▼

Provide an analysis_plan.md file containing chart specs with SQL queries and altair code. The Skill parses it, generates a <pipeline_name>_dashboard.py marimo app, validates it with uvx marimo check, and launches it with uv run marimo edit.

How to fix marimo check variable conflict errors?▼

Variable conflicts occur when two cells export the same name. Follow the naming conventions in references/notebook-patterns.md: use df_chart1, df_chart2 for dataframes and keep the altair object cell-local with the _chart underscore prefix.

Why is my marimo chart cell rendering nothing?▼

Chart cells must end with _chart on a bare line followed by return. Without the bare _chart expression line, marimo displays nothing even though the code runs without errors.

Does this work with ibis queries in marimo notebooks?▼

Yes, charts using ibis expressions are supported. Install ibis-framework[duckdb] and add import ibis inside the specific data cell that needs it rather than the global setup cell.

Why does dlt.attach fail inside the notebook?▼

dlt.attach fails when the pipeline name is wrong or the pipeline has not been run yet. Verify the pipeline exists by running dlthub local pipeline info <name> before launching the notebook.

When should I not use this notebook builder?▼

Do not use it for exploring data or planning charts, which belongs to the explore-data step, nor for building pipelines or deploying to dlthub. It only assembles dashboards from an existing analysis plan.