Does ds-brainstorm perform data exploration?

No. Following its 'Iron Law,' this skill strictly forbids data loading or exploration. Its purpose is to define the 'What' and 'Why' before moving to the 'How' in the planning phase.

What is the next step after using ds-brainstorm?

Once brainstorming is complete and the spec is written, you should move to Phase 2: ds-plan, which focuses on data profiling and task breakdown.

What is the output of a ds-brainstorm session?

The primary output is a .claude/SPEC.md file which documents the project objectives, identified data sources, success criteria, and the chosen analytical approach.

How does this skill handle existing analyses?

It specifically asks if you are replicating or extending existing work to ensure the methodology matches required standards or specific prior results.

Data Science Project Brainstormer

Name: Data Science Project Brainstormer
Author: edwinhu

byedwinhu

Ciencia de Datos y ML

Refines vague data analysis requests into clear, actionable objectives using Socratic questioning before any code is written.

Acerca de

This skill acts as the mandatory first phase of a professional data science workflow, preventing common pitfalls by enforcing a 'questions-first' approach. It guides users through a structured discovery process to identify data sources, success criteria, and constraints, specifically checking for replication requirements to ensure methodological rigor. The process concludes with the creation of a formal specification document (.claude/SPEC.md), providing a solid foundation for subsequent data profiling and implementation phases.

Características Principales

0 GitHub stars
Multi-approach proposal generation with trade-off analysis
Comprehensive data source and constraint identification
Formal replication requirement detection and documentation
Automatic generation of structured .claude/SPEC.md files
Socratic questioning framework to clarify analysis goals

Casos de Uso

Initiating a new data science project from a vague stakeholder request
Planning the extension or replication of an existing research paper or report
Defining success metrics and validation criteria before loading large datasets

Acerca de

Características Principales

0 GitHub stars
Multi-approach proposal generation with trade-off analysis
Comprehensive data source and constraint identification
Formal replication requirement detection and documentation
Automatic generation of structured .claude/SPEC.md files
Socratic questioning framework to clarify analysis goals

Casos de Uso

Initiating a new data science project from a vague stakeholder request
Planning the extension or replication of an existing research paper or report
Defining success metrics and validation criteria before loading large datasets