Project Memory FAQs

Question 1

What core files does this skill manage?

Accepted Answer

The skill manages four primary files: bugs.md (solution log), decisions.md (ADRs), key_facts.md (non-sensitive config), and issues.md (work log). It also updates CLAUDE.md and AGENTS.md to ensure AI tools proactively check these files before performing tasks.

Question 2

When should I use this skill?

Accepted Answer

Use this skill when starting a new project, documenting a recurring bug fix, or before making significant architectural changes. It is also highly effective when using multiple AI tools (like Claude Code and Cursor) to ensure they all share a single source of project truth.

Question 3

Is it safe to store credentials in Project Memory files?

Accepted Answer

No. The skill is designed to store project metadata—such as database hostnames, ports, and JIRA keys—but it includes strict guidelines to never store sensitive credentials like passwords or API keys, which should remain in your .env or secrets manager.

Question 4

What does the Project Memory skill do?

Accepted Answer

The Project Memory skill creates and maintains a structured knowledge system within your repository (stored in docs/project_notes/). it tracks bugs, architectural decisions (ADRs), key project facts, and work history, ensuring Claude remains context-aware across different coding sessions.

Question 5

How does it improve my AI-assisted development workflow?

Accepted Answer

It prevents 'AI amnesia' by providing Claude with a historical record of why decisions were made and how past problems were solved. This reduces redundant errors, speeds up onboarding, and ensures that AI-generated code aligns with your project's established standards.

Project Memory

Project Memory

概要

主な機能

ユースケース

概要

主な機能

ユースケース