automate-workflow

Analyze conversations and propose scoped automation artifacts for user approval.

Updated Sep 9, 2024
One-click install
npx skills add https://github.com/axel-kaliff/dotfiles --skill automate-workflow-axel-kaliff
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: automate-workflow
Source: https://github.com/axel-kaliff/dotfiles/tree/main/claude/.claude/skills/automate-workflow
Command: npx skills add https://github.com/axel-kaliff/dotfiles --skill automate-workflow-axel-kaliff

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill reduces repeated manual thinking by scanning a conversation for recurring patterns and translating them into deterministic automation artifacts (hooks, skills, rules, or local CLAUDE.md files) so future interactions require less friction and fewer tokens.

Core Features & Use Cases

  • Conversation analysis: Identifies repeated corrections, multi-step sequences, boilerplate choices, and verification steps from the full conversation history.
  • Scoped artifact recommendations: Classifies each pattern as user-wide, project-wide, or local and recommends the appropriate artifact type and target path.
  • Safe implementation workflow: Produces a proposal table for user approval, performs dedup checks, enforces git-exclusion for local CLAUDE.md files, and provides step-by-step implementation instructions without making changes until approved.

Quick Start

Ask the skill to analyze the current conversation and propose scoped automation artifacts with evidence and a proposal table for my approval.

Frequently Asked Questions about automate-workflow

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

FAQPage Schema
How do I turn repeated chat corrections into reusable Claude automations?

To turn repeated chat corrections into reusable Claude automations, analyze your conversation history to identify recurring patterns, then extract them into hooks, skills, rules, or CLAUDE.md files. The skill scopes each artifact as user-wide, project-wide, or directory-local.

How do I extract workflow patterns from a conversation into hooks or rules?

Extracting workflow patterns from a conversation involves scanning for multi-step sequences and boilerplate decisions, then classifying them into appropriate artifact types. A proposal table is generated requiring explicit user approval before any files are created under ~/.claude/ or .claude/.

Can I create local CLAUDE.md files without accidentally committing them to git?

Yes, you can create local CLAUDE.md files safely. The automation workflow enforces git-exclusion for local CLAUDE.md files, ensuring directory-local rules and configurations are generated without being accidentally tracked or committed to your version control system.

What is the best way to reduce manual verification steps in Claude code review sessions?

The best way to reduce manual verification steps in Claude code review sessions is to analyze the conversation for repeated corrections and translate them into deterministic automation artifacts. This reduces future friction and token usage by automating the verification process.

Does the automation proposal workflow make changes before I approve them?

No, the automation proposal workflow does not make changes before approval. It produces a proposal table with evidence and step-by-step implementation instructions, performing dedup checks and waiting for explicit user approval before creating any artifacts or updating settings.

When should I not use conversation analysis to generate Claude automation artifacts?

You should not use conversation analysis to generate automation artifacts for unique, non-recurring tasks. The process targets repeated corrections, multi-step workflows, and boilerplate decisions; applying it to one-off interactions yields low-value artifacts and unnecessary file clutter.