# Run Python Data Workflows in a Sandbox

Data and reporting tasks often need repeatable Python workflows with clear output rules. This skill guides Claude, Codex, and Claude Code through sandboxed analysis, charts, and downloadable document generation.

## Install

```bash
npx skillstore add ck991357/python-sandbox
```

## Metadata

- Status: approved
- Slug: ck991357-python-sandbox
- Version: 2.5
- Author version: 2.5
- Skillstore revision: r1
- Version status: invalid
- Tree hash: aa538221bdf0bd96e82cf3940f06c2c2afe9092e886f9d34ab9047b666e0b750
- Author: CK991357
- GitHub username: CK991357
- License: MIT
- Repository: https://github.com/CK991357/gemini-chat/tree/main/src/skills/python\_sandbox
- Ref: 7db9b9f06e0ab79c575b58bc48c4d8dc9849f424
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: scripts, network, external\_commands, filesystem
- Quality score: 75
- Public page: https://skillstore.pages.dev/skills/ck991357-python-sandbox
- Manifest: https://skillstore.pages.dev/api/skills/ck991357-python-sandbox/manifest

## Capabilities

- Explains how to read and write workspace files in a sandboxed data directory.
- Provides patterns for Pandas, DuckDB, SciPy, SymPy, and machine learning workflows.
- Guides chart creation with Matplotlib, Seaborn, Graphviz, and NetworkX.
- Shows how to generate Excel, Word, PDF, and PowerPoint outputs for download.
- Documents memory, file persistence, and multi-step session workflow practices.

## Use Cases

- Analyze Business Data: Clean spreadsheets, aggregate metrics, create charts, and export a summary workbook.
- Prototype Machine Learning Models: Prepare data, train simple models, compare metrics, and visualize model performance.
- Generate Automated Reports: Create Word, PDF, Excel, or PowerPoint reports from analysis outputs.

## Prompt Templates

### Summarize a Dataset

```
Use the Python sandbox workflow to inspect my uploaded dataset, summarize columns, identify missing values, and report key statistics.
```

### Create Charts and Insights

```
Analyze the sales data in the sandbox, create clear charts, explain the main trends, and include any data quality concerns.
```

### Build a Downloadable Report

```
Generate a downloadable business report from the analysis results, including tables, charts, and a short executive summary.
```

### Run a Multi-Step Modeling Workflow

```
Use the sandbox to clean the dataset, train a baseline model, evaluate it, tune parameters, and summarize limitations and next steps.
```

## Limitations

- Requires a compatible Python sandbox tool to execute code.
- Network access is described as disabled, so remote data fetching is not supported.
- Large files may require chunked processing to avoid memory limits.
- Generated files must follow the documented output protocol.

## Best Practices

- Keep source data in the documented workspace directory when possible.
- Use chunked processing or efficient formats for large datasets.
- Ask for charts and reports in separate steps when the analysis is complex.

## Anti Patterns

- Do not expect the sandbox to access network resources.
- Do not process sensitive files without confirming data handling requirements.
- Do not generate large files without checking memory and output limits.

## Security Audit

- Audited at: 2026-07-09T01:18:48.614\+00:00
- Summary: Static findings were reviewed against source context. The flagged patterns are documentation examples, markdown formatting, mathematical cookbook snippets, or explicit prohibited-operation warnings rather than active malicious behavior.

## Stats

- Views: 205
- Downloads: 40
- Favorites: 0
- Popularity score: 0