investigation

Scaffold dated investigation folders with README, scripts, and data.

23|16|Updated Jun 15, 2022
One-click install
npx skills add https://github.com/talmolab/sleap-io --skill investigation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigation
Source: https://github.com/talmolab/sleap-io/tree/main/.claude/skills/investigation
Command: npx skills add https://github.com/talmolab/sleap-io --skill investigation

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill scaffolds a template for a small investigation to support empirical experimentation. It helps you quickly structure a research or design-for-data process by creating a scratch workspace, a README with context, tasks, and results, and optional scripts and data to run sub-experiments.

Core Features & Use Cases

  • Structured workspace: Create a dated folder under scratch/ to host the investigation.
  • Notebook-style README: Include background notes, a task list, and a results summary.
  • Experiment scaffolding: Generate scripts and testing data files to run empirical checks.
  • Sub-experiments: For multi-step investigations, create individual markdown notes for intermediate results.
  • Actionable findings: The README should document insights to inform design decisions.

Quick Start

Create a folder in scratch/ with the format {YYYY-MM-DD}-{descriptive-name}. You will mainly work inside this folder. Then add a README.md describing the task, context, and progress, plus any scripts or data to conduct the investigation. For multi-step investigations, add intermediate notes as separate Markdown files. Ensure the README contains actionable findings to support a subsequent implementation or design decision.

Frequently Asked Questions about investigation

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

FAQPage Schema
How do I structure a small empirical investigation to inform a design decision?

Create a dated folder under scratch/ with format {YYYY-MM-DD}-{descriptive-name}. Add a README documenting task context, background, and progress, then include scripts and testing data to run experiments. This scaffold supports design decisions, performance assessment, or MVP validation with organized results and actionable findings.

What should I include in an investigation README to document my experiments?

Include background notes explaining the investigation's context, a task list with specific checks to run, results summary capturing outcomes, and actionable findings that inform subsequent decisions. For multi-step investigations, add separate markdown files for intermediate results and sub-experiment notes.

Can I use this template for validating MVP ideas or prototyping?

Yes. The investigation scaffold supports MVP validation, performance evaluation, and prototype ideation. It produces a workspace with supporting scripts and testing data, letting you run empirical checks and document insights to validate assumptions before full implementation.

How do I organize multi-step experiments within a single investigation?

Structure each experiment step as a separate markdown note file within your dated investigation folder. Document intermediate results and findings in each note, then synthesize conclusions in the main README to track progress across the full investigation.

What format should investigation folders follow to stay organized?

Use {YYYY-MM-DD}-{descriptive-name} as your folder name under scratch/. This date-prefixed, descriptive naming keeps investigations chronologically sortable and quickly identifiable by purpose, supporting efficient workspace navigation.

When should I use an investigation scaffold instead of starting code directly?

Use this scaffold when you need empirical data to inform decisions, evaluate performance, or prototype ideas before committing to implementation. It structures research and experimentation work, reducing uncertainty through documentation and systematic testing.