ds-subagent-driven-analysis

Community

Execute plan tasks fast with safer review.

AuthorKhodzitcky-Vl
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill reduces the risk of incorrect, non-reproducible analytics by executing each step of an analysis plan with the right review depth and escalation on validation “risk signals.”

Core Features & Use Cases

  • Per-task subagent dispatch: Runs independent task scopes using a fresh subagent for each analysis task, preserving plan context rather than reusing a potentially stale workflow.
  • Validation-budgeted methodology review: Escalates to a methodology compliance reviewer for standard/strict tasks and for tasks that impact final conclusions or trigger risk signals.
  • Reproducibility and interpretation checking: Escalates to a dedicated reproducibility/interpretation reviewer after methodology passes to confirm reruns, parameter clarity, notebook self-containment, and that conclusions match outputs.
  • Notebook execution hygiene: Maintains runtime observability without polluting saved notebook outputs with progress-spam.

Quick Start

Use ds-subagent-driven-analysis to execute the tasks from your existing ds-analysis-plan in the current session, assigning a fresh worker per task and triggering methodology plus reproducibility reviews for standard/strict and risk-sensitive steps.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: ds-subagent-driven-analysis
Download link: https://github.com/Khodzitcky-Vl/data-science-ai-superpowers/archive/main.zip#ds-subagent-driven-analysis

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.