memory-remember

Store conversation memories into structured files using memory-settings.json.

6|Updated Oct 16, 2025
One-click install
npx skills add https://github.com/legacybridge-tech/claude-plugins --skill memory-remember
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-remember
Source: https://github.com/legacybridge-tech/claude-plugins/tree/main/engram/skills/memory-remember
Command: npx skills add https://github.com/legacybridge-tech/claude-plugins --skill memory-remember

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automatically stores salient information from conversations and events into structured memory files, ensuring important context is retained across sessions.

Core Features & Use Cases

  • Auto-triggered memory saving when new preferences, decisions, milestones, or a conversation ends.
  • Classifies content into Preferences, Conversations, and Long-Term Memory, routing it to the appropriate memory file.
  • Reads .claude/memory-settings.json to determine file names and placement rules; writes non-destructively (preferences: update sections and Last Updated date; conversations: prepend to the top; long-term memory: append under milestones, decisions, or lessons learned).
  • Example: after a planning session, memory remembers decisions and milestones for future reference.

Quick Start

Remember a memory using the Engram workflow, e.g., Remember: user prefers dark mode.

Frequently Asked Questions about memory-remember

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

FAQPage Schema
How do I automatically remember and organize key memories from conversations across sessions?

This Skill automatically saves key memories by classifying conversation content into preferences, conversations, and long-term milestones, then routing them to structured memory files for persistent recall across sessions.

How does knowledge-management classification route memories to different files?

Knowledge-management classification routes memories by category: preferences update existing sections with a Last Updated date, conversations prepend to the top of the file, and long-term memory appends under milestones, decisions, or lessons learned.

Do I need a memory-settings.json file to configure where memories are stored?

Yes, you need a memory-settings.json file. The Skill reads this configuration file to determine specific memory file names and placement rules before writing any classified memories non-destructively.

What is the best way to ensure non-destructive writes when saving long-term milestones?

The best way to ensure non-destructive writes for long-term milestones is appending new entries under existing milestones, decisions, or lessons learned sections, which preserves prior context while adding new information.

When do memories get auto-triggered for saving during a workflow?

Memories are auto-triggered for saving when new preferences, decisions, or milestones are detected during a conversation, or when a conversation ends, capturing salient information without manual intervention.