skills.rest
jubalm
jubalmCOMMUNITY
Explore all skills by this author →
GitHub

claude-code-memory

Manage Claude Code project memory with structured files and Python utilities.

1|Updated Oct 1, 2023
Search Tags:#automation#memory#inbox#CLAUDE.md#token-efficiency#memory-audit
One-click install
npx skills add https://github.com/jubalm/dotfiles --skill claude-code-memory-jubalm
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-code-memory
Source: https://github.com/jubalm/dotfiles/tree/main/home/.claude/skills/claude-code-memory
Command: npx skills add https://github.com/jubalm/dotfiles --skill claude-code-memory-jubalm

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a structured system for capturing and organizing project-specific memory to enhance Claude Code assistance. Memory is auto-loaded into context via CLAUDE.md, while an Inbox stores uncertainties for later review, and an audit system helps detect bloat and conflicts.

Core Features & Use Cases

  • Memory structure under .claude/memory with constraints.md, quirks.md, decisions.md, conventions.md
  • CLAUDE.md imports and auto-load behavior
  • Inbox for uncertainties to avoid polluting memory
  • Audit & promote: audit memory and promote inbox items when confirmed
  • Manual entry & workflow: add entries and manage conflict resolution

Quick Start

Initialize the memory system: run scripts/init_memory.py to create the structure, then add an inbox item with scripts/inbox_add.py, and finally review inbox items with scripts/inbox_list.py.

Frequently Asked Questions about claude-code-memory

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

FAQPage Schema
How do I set up project memory for Claude Code to reduce token usage?▼

Project memory reduces token overhead by storing constraints, quirks, decisions, and conventions in structured files under .claude/memory, then auto-loading them via CLAUDE.md. Run scripts/init_memory.py to initialize the structure, then add entries using scripts/inbox_add.py and promote confirmed items from inbox to permanent memory files.

What's the difference between memory files and the inbox in Claude Code projects?▼

Memory files (constraints.md, quirks.md, decisions.md, conventions.md) hold confirmed, reusable project knowledge that auto-loads into context. The inbox collects uncertainties and unvalidated observations with frontmatter metadata, letting you defer decisions until patterns emerge, then selectively promote inbox items to memory when verified.

How do I audit Claude Code project memory for conflicts and redundancy?▼

Use scripts/audit_memory.py to scan memory files for bloat, duplication, and conflicting entries. The audit detects when the same constraint appears across multiple files or when decisions contradict conventions, helping you consolidate and keep memory lean and coherent.

Can I capture project uncertainties without polluting my main memory?▼

Yes—the inbox system captures uncertainties, edge cases, and tentative notes separately from confirmed memory. Add items with scripts/inbox_add.py, review with scripts/inbox_list.py and scripts/inbox_show.py, then promote only validated entries to permanent memory using scripts/inbox_promote.py.

Do I need Python to manage Claude Code project memory?▼

Yes—the memory system includes Python utility scripts (init_memory.py, add_entry.py, audit_memory.py, inbox_add.py, inbox_promote.py, inbox_list.py, inbox_show.py) to initialize, add, audit, and manage memory and inbox items. These scripts automate deduplication, metadata handling, and conflict detection.

How does CLAUDE.md auto-load project memory into Claude Code context?▼

CLAUDE.md imports memory files from .claude/memory during context initialization, automatically injecting constraints, quirks, decisions, and conventions into every Claude Code task without manual copying. This keeps project context consistent and token-efficient across sessions.

Related Skills

Explore More Software Engineering Skills→
stratusstratus
2

remember

Persist user learnings and preferences across sessions in local memory files.

Community
Advanced
RwanbtRwanbt
1

verify-ai-docs

Audit AI documentation, dependency graphs, and memory synchronization across project structures.

Community
Advanced
mhattingpetemhattingpete
657

conversation-analyzer

Analyze Claude Code conversation history to identify usage patterns and automation opportunities.

Community
Advanced
lobehublobehub
81.4k

microcopy

Guide UI writers to produce consistent bilingual microcopy in English and Chinese.

Official
Basic