domain-memory

Manage persistent agent state in domain-memory.yaml using bash scripts.

Updated Jan 5, 2026
One-click install
npx skills add https://github.com/Marcus-Mok-GH/Claude-code-conductor --skill domain-memory-marcus-mok-gh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: domain-memory
Source: https://github.com/Marcus-Mok-GH/Claude-code-conductor/tree/main/.claude/skills/domain-memory
Command: npx skills add https://github.com/Marcus-Mok-GH/Claude-code-conductor --skill domain-memory-marcus-mok-gh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

The Domain Memory Skill provides a standardized, disk-based approach to state management for stateless agents, ensuring persistent context and observable progress in development workflows.

Core Features & Use Cases

  • Persistent State Management: Store all agent state in domain-memory.yaml, ensuring no data loss during restarts.
  • Disk-Based Operations: Facilitates stateless agent workflows by having agents read and write directly to the disk.
  • Feature Status Tracking: Track the progress of features through defined states like untested, passing, failing, in_progress, and blocked.
  • Integration with Claude Code: Works seamlessly with Claude Code's slash commands for workflow management.

Quick Start

Initialize the domain-memory for a new spec: domain-memory.sh init specs/001-auth

Frequently Asked Questions about domain-memory

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

FAQPage Schema
How do I maintain persistent workflow state for stateless agent workflows?

Track feature progress in development workflows by saving status states like passing, failing, in_progress, or blocked directly to disk, allowing stateless agents to retrieve current task context upon initialization.

How do I initialize persistent state management for a new development spec?

To initialize persistent state management for a new spec, run the bash initialization script targeting your spec directory, which sets up the required YAML file for tracking development task progress and state.

Do I need external dependencies to manage agent state with bash scripts?

No, you do not need external dependencies to manage agent state with bash scripts, because the implementation handles state initialization, reading, updating, and logging entirely through built-in shell operations.

Can I integrate disk-based state management with Claude Code slash commands?

Yes, you can integrate disk-based state management with Claude Code slash commands, because the YAML state file works seamlessly with Claude Code's workflow management to update and log task progress.

What are the limitations of using disk-based YAML for agent state management?

A limitation of using disk-based YAML for agent state management is that state concurrency relies on file I/O operations, meaning simultaneous write conflicts may occur if multiple stateless agents attempt to update the same file without locking.