skillify

Captures a session's repeatable workflow into a reusable SKILL.md file.

2|Updated Oct 16, 2021
One-click install
npx skills add https://github.com/pattobrien/dotfiles --skill skillify-pattobrien
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skillify
Source: https://github.com/pattobrien/dotfiles/tree/main/claude/skills/skillify
Command: npx skills add https://github.com/pattobrien/dotfiles --skill skillify-pattobrien

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After completing a useful multi-step process in a Claude session, the workflow is lost unless manually documented. This Skill analyzes the conversation history and interviews the user to convert that session into a reusable, invocable skill. ## Core Features & Use Cases - Session Analysis: Reviews the full conversation history to identify steps, tools, corrections, and success criteria from the completed process. - Structured Interview: Uses AskUserQuestion rounds to confirm the skill name, goals, arguments, execution mode (inline or forked), and save location. - SKILL.md Generation: Writes a properly formatted SKILL.md with frontmatter (name, description, when_to_use, allowed-tools, arguments) and per-step success criteria, artifacts, and human checkpoints. - Use Case: After spending a session cherry-picking a PR to a release branch with CI monitoring, invoke /skillify to capture that exact workflow so it can be rerun with a single command next time. ## Quick Start Run /skillify with a short description of the process you just completed to turn this session into a reusable skill.

Frequently Asked Questions about skillify

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

FAQPage Schema
How do I turn a Claude session into a reusable skill?▼

Invoke /skillify at the end of the session, optionally with a description of the process. It analyzes the conversation history, interviews you to confirm steps and goals, then writes a SKILL.md file to your chosen location.

Where should a generated skill be saved?▼

Repo-specific workflows go in .claude/skills/<name>/SKILL.md within the project, while cross-repo personal workflows go in ~/.claude/skills/<name>/SKILL.md so they follow you across all repositories.

What is the difference between inline and forked skill execution?▼

Inline skills run in the current conversation and suit workflows where you want to steer mid-process. Forked skills run as a sub-agent with their own context, which works better for self-contained tasks needing no mid-process input.

What frontmatter fields does a generated SKILL.md include?▼

Generated skills include name, description, when_to_use trigger phrases, allowed-tools permission patterns, and optionally argument-hint, arguments, and context. The when_to_use field is critical because it tells the model when to auto-invoke the skill.

When should I not use skillify to capture a workflow?▼

Avoid it for one-off tasks that will not repeat or trivially simple processes, since the interview rounds add overhead. It is designed for multi-step workflows with clear success criteria that you expect to perform again.