Jupyter
DevelopmentjupyterJupyter is an Agent Skill for development published in the Prismer-AI/Prismer repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Create and execute Jupyter notebooks for interactive data analysis using jupyter_execute and jupyter_notebook tools. Use when the user asks to run Python code interactively, create notebooks, analyze data in cells, or mentions .ipynb files.
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/Prismer-AI/Prismer/blob/HEAD/docker/templates/data-scientist/skills/jupyter/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/jupyter, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
Jupyter Notebook Skill
Description
Create and execute Jupyter notebooks for interactive data analysis and visualization.
Tools Used
jupyter_execute- Execute Python code in Jupyter kernel (auto-switches to Jupyter)jupyter_notebook- Create, read, update, delete, and list notebooksupdate_notebook- Add or update cells in the notebook without executingupdate_gallery- Display generated plots and visualizations in gallery viewupdate_data_grid- Display structured tabular data (DataFrames, query results) in AG Gridupdate_code- Show code examples and scripts in the Code Playgroundsave_artifact- Save generated artifacts (plots, data files) to workspace collection
Capabilities
- Create new notebooks with proper structure
- Add and execute code cells
- Add markdown documentation cells
- Display inline visualizations
- Display tabular data in interactive grid view
- Show code examples with syntax highlighting
- Export to various formats (HTML, PDF)
Usage Patterns
Create Analysis Notebook
When user says: "Create a notebook for [analysis]"
- Create notebook with title and imports
- Add data loading cell
- Add exploration cells
- Structure with markdown headers
- Execute cells sequentially
Execute and Debug
When user says: "Run this code"
- Execute cell
- Capture output and errors
- If error, diagnose and fix
- Show results or visualizations
Document Workflow
When user says: "Add explanation for this step"
- Add markdown cell before code
- Explain methodology
- Note assumptions and limitations
Tool Examples
Execute Python code
jupyter_execute code="import pandas as pd\ndf = pd.read_csv('/workspace/data/results.csv')\nprint(df.describe())"
Create a notebook with cells
update_notebook cells=[{"type": "markdown", "source": "# Analysis"}, {"type": "code", "source": "import pandas as pd\nimport matplotlib.pyplot as plt"}] execute=false
Display generated plots
update_gallery images=[{"url": "/workspace/data/plot.png", "title": "Analysis Results"}]
Best Practices
- Cell Independence: Each cell should run independently when possible
- Import First: All imports at notebook start
- Check output before proceeding: Verify execution output is correct before running dependent cells
- Markdown Structure: Use headers for navigation
- Save Often: Checkpoint regularly