Workflow Skill Creator

This turns workflows you've already completed into reusable skills that Claude can apply elsewhere. The key insight here is distillation: instead of manually documenting a pattern, you let it extract the structure from an actual interaction. The mandatory brainstorming phase (Phase 1) is non-negotiable, which feels heavy handed until you realize most people skip straight to writing code and end up with brittle abstractions. From Google DeepMind's science-skills collection, so it's built for research workflows where you're doing something novel once, then want to templatize it. Worth trying if you find yourself repeatedly copying and tweaking old conversation patterns instead of properly capturing what made them work.

skills.sh installs: 2.5K
GitHub stars: 3.4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add google-deepmind/science-skills --skill workflow-skill-creator --agent claude-code

Installs into .claude/skills of the current project.

Files

Workflow-to-Skill Distiller

Turns a completed workflow into a reusable agent skill. Specifically, this skill extracts patterns from an interaction or workflow that already happened and packages them.

[!CAUTION] You MUST complete Phase 1 (Brainstorming) before writing any code or SKILL.md content. Skipping brainstorming produces skills that are either too rigid or too vague. The brainstorming conversation is the most important part of this process.

Phase 1: Brainstorming (MANDATORY)

Have an iterative back-and-forth conversation with the user. Do NOT ask all questions at once. Pick 2-3 relevant questions per round from the bank below, refine your understanding, and ask follow-ups.

Round 1: Understand the Workflow

Start by summarizing what you observed from the workflow, then ask:

"Here's my understanding of the workflow: [summary]. Is this accurate? What would you change?"
"What are the expected inputs and outputs for this workflow?"
"How often do you expect to run this workflow? Is it recurring or one-off?"