What is the recall-memory skill for Claude Code?

It is a skill within the memU plugin that allows Claude to access and retrieve information from previous conversations, including your preferences, coding patterns, and past discussions.

Can I retrieve specific coding styles I mentioned in previous weeks?

Yes, as long as the information was discussed in a session managed by memU, you can query it using keywords or descriptive questions like 'What patterns do I use for React components?'

How does memU organize stored information?

memU uses a three-layer hierarchy: Resources (raw session data), Items (specific facts, habits, and opinions), and Categories (high-level summaries of your knowledge).

When does Claude save information to my memory?

Conversations are automatically processed and memorized by the memU framework at the end of each session via the SessionEnd hook.

When should I use RAG vs. LLM retrieval methods?

Use RAG (Retrieval-Augmented Generation) for fast, factual queries where you know what you're looking for. Use the LLM method for complex, nuanced, or multi-faceted questions that require deep semantic understanding.

Recall Memory (memU)

Name: Recall Memory (memU)
Author: asermax

byasermax

•

Productividad y Flujo de Trabajo

Retrieves historical context, user preferences, and coding patterns from past Claude Code sessions to provide personalized assistance.

Acerca de

Recall Memory is a specialized Claude Code skill that leverages the memU agentic memory framework to provide Claude with persistent, long-term memory across sessions. By accessing a structured three-layer hierarchy—consisting of raw conversation resources, extracted facts/preferences, and aggregated summaries—this skill allows Claude to remember your specific coding styles, project decisions, and workflows. Whether you need a quick factual check via RAG or a deep semantic analysis of your habits via LLM-based retrieval, this skill ensures your AI assistant evolves with you, eliminating the need to repeat instructions in every new session.

Características Principales

JSON-structured output for seamless integration into AI workflows
Three-layer memory architecture: Resources, Items, and Categories
Context-aware retrieval of coding patterns and development habits
Dual-method retrieval using RAG for speed and LLM for semantic depth
Automatic cross-session knowledge persistence and traceability
1 GitHub stars

Casos de Uso

Accessing personalized user preferences to maintain consistency across sessions
Retrieving previously discussed debugging steps or environment configurations
Recalling project-specific architectural patterns and naming conventions

Acerca de

Características Principales

JSON-structured output for seamless integration into AI workflows
Three-layer memory architecture: Resources, Items, and Categories
Context-aware retrieval of coding patterns and development habits
Dual-method retrieval using RAG for speed and LLM for semantic depth
Automatic cross-session knowledge persistence and traceability
1 GitHub stars

Casos de Uso

Accessing personalized user preferences to maintain consistency across sessions
Retrieving previously discussed debugging steps or environment configurations
Recalling project-specific architectural patterns and naming conventions