Memory

Persist and retrieve atomic memories with tags across sessions.

3|1|Updated Jun 23, 2024
One-click install
npx skills add https://github.com/channingwalton/dotfiles --skill memory-channingwalton
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Memory
Source: https://github.com/channingwalton/dotfiles/tree/main/.claude/skills/memory
Command: npx skills add https://github.com/channingwalton/dotfiles --skill memory-channingwalton

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you overcome the challenge of remembering and recalling crucial information across tasks and sessions, preventing repetitive searches and ensuring consistency in your work.

Core Features & Use Cases

  • Intelligent Recall: Seamlessly search existing memory for relevant past observations, entity names, and types.
  • Structured Knowledge Storage: Store atomic facts with UTC timestamps and establish entity relationships (e.g., depends-on, solves) for precise future retrieval.
  • Use Case: Before tackling a complex coding task or project, quickly search the memory for previously recorded solutions, architectural decisions, or specific entity details to leverage prior knowledge and accelerate your progress.

Quick Start

Search existing memory for "foggysky-authentication setup" to retrieve relevant project information.

Frequently Asked Questions about Memory

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

FAQPage Schema
How do I persist and retrieve information across different sessions and projects?

Persist and retrieve information using the Memory Skill's retrieve_memory, recall_memory, and search_by_tag functions. Store atomic facts with descriptive tags and UTC timestamps to maintain context across sessions, projects, and topics, preventing repetitive searches and ensuring consistency in your work.

What's the best way to structure knowledge storage for accurate information retrieval?

Structure knowledge storage by storing one fact per memory with descriptive tags and establishing entity relationships using tags like depends-on and solves. This atomic approach enables precise future retrieval and seamless searching of past observations, entity names, and architectural decisions.

How do I search existing memory before starting a complex coding task?

Search existing memory at session start using search_by_tag and recall_memory to retrieve previously recorded solutions, architectural decisions, and entity details. This retrieval accelerates progress on complex tasks by leveraging prior knowledge and avoiding duplicate work across projects.

Can I use Memory to track relationships between different entities and concepts?

Yes, establish entity relationships through descriptive tags that define connections like depends-on and solves. The MCP-Memory Server backing enables storing structured metadata alongside content, allowing you to encode and retrieve complex relationships between entities and concepts.

What happens when I store learnings after task completion?

Store learnings after task completion to populate your knowledge base with atomic facts tagged for searchability. Automatic storage triggers persist facts with UTC timestamps and metadata, making them available for future retrieval across all sessions and projects.