Init Harness FAQs

Question 1

What does the Init Harness skill do?

Accepted Answer

Init Harness establishes a structured task management framework for AI agents. it automatically generates feature backlogs, session audit logs, and standardized instruction files (CLAUDE.md) while detecting your project's architecture.

Question 2

How does this skill improve my AI coding workflow?

Accepted Answer

It eliminates manual setup by auto-scanning your source code and creating a clear roadmap for the AI. By providing a progress log and feature list, it ensures the AI remains focused, organized, and accountable for every task.

Question 3

What files are created by the agent-foreman init command?

Accepted Answer

The command creates an 'ai/' directory containing feature_list.json, progress.log, and init.sh. It also generates a CLAUDE.md file for agent instructions and an ARCHITECTURE.md file to help the AI understand your project structure.

Question 4

What are the different TDD modes available?

Accepted Answer

Init Harness offers three modes: Strict (tests are required for completion), Recommended (tests are suggested but optional), and Disabled (no TDD guidance provided). This allows you to customize the level of oversight required for your project.

Question 5

When should I use the Init Harness skill?

Accepted Answer

You should use this skill when starting a new development project, onboarding an AI agent to an existing codebase, or when you need to enforce strict Test-Driven Development (TDD) practices for AI-generated code.

Init Harness

Init Harness

概要

主な機能

ユースケース

概要

主な機能

ユースケース