regardo911

Seven Research Workflows — Development skill for Claude Code

Development community

Seven Claude research workflows and the verification layer that proves each one, as files you own.

How to install Seven Research Workflows

This entry records only its repository, not the path inside it, so there is no exact command to give. Open regardo911/seven-research-workflows and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Seven Research Workflows does

Seven Claude research workflows and the verification layer that proves each one, as files you own. Companion resources to the book. Educational software, not research or regulatory advice - from youcanbuildthings.com.

Alternatives in Development

  • Book To Skill — Turn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work 25.9k ★
  • Claude Scholar — Semi-automated research assistant for academic research and software development 3k ★
  • Analyze Codebase To Book — Prompt: Turn Any Codebase Into a Technical Book 1.7k ★

README

seven-research-workflows

**Seven Claude research workflows and the verification layer that proves each one, as files you own.** Companion repository to the book *Claude for Life Sciences: Build 7 Research Workflows*, from [youcanbuildthings.com](https://youcanbuildthings.com).

In plain words: this is a folder of instruction files and small scripts that make an AI assistant check its own work against facts on disk, so your results survive somebody asking where a number came from. Two words you will meet in a minute, *Skill* and *rubric*, are just a markdown file with rules in it and a checklist whose every row resolves against a file. You don't need to know either one yet. Start below, at whichever line describes you.

![The chapter-to-chapter artifact chain. Chapter 2's trust-boundary.md is copied into the reviewer Skill's references directory in Chapter 8. Chapter 5's 00_clean.py produces final_table.tsv and drop_log.tsv, which verify.sh reads for rows 3 and 4, and which go.sh runs in Chapter 10. Chapter 6's config.yaml travels into every run directory as config.used.yaml and supplies the numbers for the Chapter 9 methods paragraph. Chapter 7's conda-lock.yml is what the reproduction claim rests on. Every arrow is one file being handed from the chapter that makes it to the tool that reads it.](images/hero.png)

Start here

**I finished the book and want the files.**

mkdir -p ~/.claude/skills && cp -r chapters/*/skills/* ~/.claude/skills/ && bash validate-all.sh ~/.claude/skills
Valid skill: /Users/you/.claude/skills/citation-check
Valid skill: /Users/you/.claude/skills/data-cleaning-verify
Valid skill: /Users/you/.claude/skills/reviewer
Valid skill: /Users/you/.claude/skills/samplesheet-audit
4 Skills checked, all valid

Four directories land in `~/.claude/skills/` and Claude Code loads them in every session from now on. If you already have a Skill by one of those names, copy them one at a time instead. `pip install skills-ref` first if