hac-init

Create the .hac directory structure with HAC files and update CLAUDE.md.

2|1|Updated Apr 22, 2026
One-click install
npx skills add https://github.com/pytholic/claude-skills --skill hac-init
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hac-init
Source: https://github.com/pytholic/claude-skills/tree/main/hac-init
Command: npx skills add https://github.com/pytholic/claude-skills --skill hac-init

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill sets up a lightweight Human-Agent Context workspace so a project can be resumed quickly without digging through chat history, old notes, or scattered TODO files.

Core Features & Use Cases

  • Project memory bootstrap: Creates the .hac/ structure for status tracking, decisions, and per-task scratchpads.
  • Session continuity: Helps humans and agents pick up multi-session work with clear context in under 30 seconds.
  • Decision capture: Records architectural or strategic choices in an append-only log so rationale stays visible later.
  • Task tracking: Organizes active, blocked, parked, and completed work around a simple project dashboard.
  • Use case: A team starting a multi-step refactor can initialize HAC, log the plan, and resume later without re-explaining the project.

Quick Start

Ask Claude to set up HAC for the project and add the .hac/ context files.

Frequently Asked Questions about hac-init

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

FAQPage Schema
How do I set up shared project context for multi-session AI development?

To set up shared project context for multi-session AI development, initialize a .hac directory structure. This creates standard files for status tracking, decision logging, and task handoff, updating CLAUDE.md with the protocol to preserve persistent context across human and AI collaborators.

What's the best way to track project decisions and rationale across chat sessions?

Tracking project decisions across chat sessions requires an append-only decision log. By recording architectural and strategic choices in a dedicated file within the project context directory, the rationale stays visible and accessible later without digging through scattered notes or chat history.

How do I resume a complex coding task with Claude without re-explaining the project?

To resume a complex coding task without re-explaining the project, use a shared context workspace with a project dashboard. This organizes active, blocked, parked, and completed work, allowing humans and agents to pick up multi-session work with clear context in under 30 seconds.

Can I use persistent task tracking for multi-step refactoring with AI collaborators?

Yes, persistent task tracking supports multi-step refactoring with AI collaborators. You can initialize a lightweight workspace, log the refactoring plan, and organize per-task scratchpads. This allows the team to resume work later without re-explaining the project structure or goals.

Does bootstrapping shared project memory require external dependencies or frameworks?

Bootstrapping shared project memory requires no external dependencies or frameworks. It creates a standard directory structure with context files and updates CLAUDE.md directly, keeping the setup lightweight and self-contained within the existing project repository.

When do I need an append-only decision history for my development workflow?

You need an append-only decision history when working on multi-session development that involves architectural or strategic choices. It records rationale in a persistent log so context remains visible to both human and AI collaborators, preventing knowledge loss during task handoffs.