update-memory

Capture end-of-session insights and persist them to the per-project memory store.

Updated May 14, 2026
One-click install
npx skills add https://github.com/jaqg/daimon --skill update-memory-jaqg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: update-memory
Source: https://github.com/jaqg/daimon/tree/main/vault/skills/update-memory
Command: npx skills add https://github.com/jaqg/daimon --skill update-memory-jaqg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

End-of-session memory capture for projects to preserve insights, decisions, and references across future interactions with Claude.

Core Features & Use Cases

  • Capture project facts, user corrections, and validated approaches from the current session.
  • Create or update memory files under the project memory directory with one-line summaries and metadata.
  • Use across projects to bootstrap context in new sessions.

Quick Start

Save the current session memory by issuing the update-memory command.

Frequently Asked Questions about update-memory

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

FAQPage Schema
How do I persist project memory and decisions across Claude sessions?

You persist project memory across Claude sessions by capturing end-of-session insights, decisions, and references into memory files with YAML frontmatter. These files are indexed using a MEMORY.md file for durable context bootstrapping in future interactions.

What is the best way to capture end-of-session notes for durable project context?

The best way to capture end-of-session notes is to save validated approaches and user corrections as memory files under a per-project memory directory. Each file includes a one-line summary, metadata, and proper indexing within MEMORY.md.

Do I need a specific directory setup to store project memory files?

Yes, storing project memory files requires a MEMORY_DIR derived path. This per-project memory directory houses the individual memory files with YAML frontmatter and the MEMORY.md index needed for proper context retrieval.

How does indexing work for session memory files?

Indexing session memory files works by updating a central MEMORY.md file within the memory directory. This index references the individual memory files containing YAML frontmatter, ensuring retained context, facts, and references are discoverable in new sessions.

Can I use this session memory capture across multiple different projects?

Yes, you can use this session memory capture across any project. It is designed to be applicable wherever durable context, project decisions, and validated references need to be retained for future Claude interactions.

What format do memory files need to be in for session context retention?

Memory files must use YAML frontmatter including name, description, and metadata fields for session context retention. This structured format allows the stored project facts and insights to be properly indexed in MEMORY.md.