Can I skip the review step when defining a new job?

While the workflow follows a sequence (define > review > implement), the review step is highly recommended as it uses a sub-agent to ensure your job specification meets quality criteria before implementation.

What is the purpose of the DeepWork Jobs skill?

DeepWork Jobs is designed to help you define, implement, and manage complex, multi-step AI workflows within Claude Code, ensuring consistency and efficiency in repetitive tasks.

Where are the job definitions stored?

Job definitions and specifications are stored within your project's working directory under the .deepwork/jobs/ path.

How does the 'learn' command improve workflows?

The learn command analyzes past conversation history to find points of confusion or inefficiency, then automatically updates instructions and agent files to prevent those issues in future runs.

DeepWork Jobs

Name: DeepWork Jobs
Author: Unsupervisedcom

byUnsupervisedcom

•

Productivity & Workflow

Orchestrates complex multi-step AI workflows through interactive definition, automated implementation, and continuous learning.

About

DeepWork Jobs is a powerful framework for Claude Code designed to manage sophisticated, multi-stage AI workflows from conception to execution. It simplifies the development process by guiding users through an interactive definition phase, validating job specifications with sub-agent reviews, and automatically generating implementation files. Beyond execution, the skill features a unique learning mechanism that analyzes previous conversation history to identify inefficiencies and refine job instructions, ensuring your AI workflows become more accurate and effective over time.

Key Features

5 GitHub stars
One-click generation of step instructions and slash commands
Standardized multi-step execution logic with built-in guardrails
Interactive workflow definition through structured questioning
Automated job specification review using validation sub-agents
Self-improving 'learn' command to optimize workflows from history

Use Cases

Defining and automating complex code refactoring pipelines
Creating self-correcting documentation generation workflows
Managing multi-stage development tasks that require consistent quality checks

About

Key Features

5 GitHub stars
One-click generation of step instructions and slash commands
Standardized multi-step execution logic with built-in guardrails
Interactive workflow definition through structured questioning
Automated job specification review using validation sub-agents
Self-improving 'learn' command to optimize workflows from history

Use Cases

Defining and automating complex code refactoring pipelines
Creating self-correcting documentation generation workflows
Managing multi-stage development tasks that require consistent quality checks