skills.rest
l0lxl0lw
l0lxl0lwCOMMUNITY
Explore all skills by this author →
GitHub

save-memory

Persists session knowledge into a structured MEMORY.md file at the repository root.

Updated Dec 18, 2025
Search Tags:#workflow#markdown#context management#claude#memory#knowledge capture#session persistence
One-click install
npx skills add https://github.com/l0lxl0lw/dotfiles --skill save-memory-l0lxl0lw
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: save-memory
Source: https://github.com/l0lxl0lw/dotfiles/tree/main/ai/shared/skills/utilities/save-memory
Command: npx skills add https://github.com/l0lxl0lw/dotfiles --skill save-memory-l0lxl0lw

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Long chat sessions accumulate valuable context—decisions, findings, file paths, and preferences—that disappears when the session ends or context overflows. This Skill captures that working memory into a persistent MEMORY.md file so future sessions can restore it. ## Core Features & Use Cases - Session Knowledge Extraction: Reviews the entire conversation and organizes findings into sections like Current State, Decisions Made, Key Files, and Next Steps. - Merge-Based Persistence: Reads any existing MEMORY.md and merges new information instead of overwriting, keeping entries dated and concise. - CLAUDE.md Coordination: Avoids duplicating CLAUDE.md content and suggests promoting stable, recurring patterns into permanent project instructions. - Use Case: Before ending a long debugging or research session, invoke the skill to write a structured MEMORY.md at the repo root, then restore it later with /load-memory. ## Quick Start Ask the AI to save everything from this session into MEMORY.md at the repository root before ending the conversation.

Frequently Asked Questions about save-memory

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

FAQPage Schema
How do I save chat session context before it runs out?▼

Invoke the save-memory skill before context overflows or before ending a long session. It reviews the full conversation and writes topics, decisions, file paths, and next steps into MEMORY.md at the repository root.

How do I restore saved memory in a new session?▼

The skill writes MEMORY.md at the repo root next to CLAUDE.md, designed to be restored in a future session with the companion /load-memory command. The file is structured so a session with zero prior context can pick up the work.

Does saving memory overwrite my existing MEMORY.md file?▼

No. The skill always reads the existing MEMORY.md first and merges new information into it, updating outdated sections rather than keeping stale content. It never overwrites the file blindly.

What is the difference between MEMORY.md and CLAUDE.md?▼

MEMORY.md stores session-specific working state like progress, decisions, and next steps, while CLAUDE.md holds permanent project instructions. The skill avoids duplicating CLAUDE.md content and asks before promoting stable patterns into it.

When should I not use session memory saving?▼

Skip it for short, one-off conversations with no lasting decisions or unfinished work. The skill is intended for substantive sessions where findings, preferences, or next steps would otherwise be lost.

Related Skills

Explore More Software Engineering Skills→
robritacca-dotcomrobritacca-dotcom
1

land

Triages and merges all pending git branches, worktrees, and stashes into a local main.

Community
Advanced
LiquilabLiquilab

continue

Restore context from session files and MEMORY.md to resume work.

Community
Intermediate
alirezarezvanialirezarezvani
25.3k

promote

Promotes learned patterns from auto-memory MEMORY.md into CLAUDE.md or .claude/rules/ as enforced instructions.

Community
Intermediate
travisslukatravissluka

jedi-prs

Coordinate multi-repo pull requests across JCSDA-internal repositories with merge-order reasoning.

Community
Advanced