memory-retrieve

Score MEMORY.md entries by tag overlap, keyword match, recency, and importance.

7|1|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/Shiien/Self-Evolved-Research-Framework --skill memory-retrieve-shiien
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-retrieve
Source: https://github.com/Shiien/Self-Evolved-Research-Framework/tree/main/skills/memory-retrieve
Command: npx skills add https://github.com/Shiien/Self-Evolved-Research-Framework --skill memory-retrieve-shiien

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Surface relevant persisted memories (episodes, topics, procedures) into the current conversation by scoring MEMORY.md entries on tag overlap, keyword match, recency, and importance.

Core Features & Use Cases

  • Surface top memory candidates for grounding the current task.
  • Score and filter memories by tag overlap, keyword match, recency, and importance.
  • Load top-3 memories into the session context for subsequent skills like plan-suggest or status-report.

Quick Start

Provide the most relevant memories from MEMORY.md into the current session context to ground the task.

Frequently Asked Questions about memory-retrieve

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

FAQPage Schema
How do I retrieve and surface persisted memories to ground a conversation?

Memory retrieval scores persisted entries using tag overlap, keyword matching, recency, and importance to surface relevant episodes, topics, and procedures into the active context.

How does memory scoring by recency and importance work?

Scoring persisted memories involves calculating tag overlap and keyword matches, then weighting candidates by recency and importance to filter and rank top memory results.

When do I need to load persisted memories into session context?

Load persisted memories at session-open and before executing knowledge-dependent skills to ground the current task with relevant episodes, topics, and procedures.

Can I use conversation memory retrieval without any external dependencies?

Yes, conversation memory retrieval operates without external dependencies by reading the local MEMORY.md file and processing entries internally to surface top memory candidates.

What is the best way to filter MEMORY.md entries for relevant context?

Filter MEMORY.md entries by applying a multi-factor scoring system evaluating tag overlap, keyword match, recency, and importance to load only the top-3 memories into session context.