session-memory

Persist project memory across Claude Code sessions using INDEX.md and session logs.

Updated May 12, 2026
One-click install
npx skills add https://github.com/RorySullivan1/claudeBrain --skill session-memory-rorysullivan1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-memory
Source: https://github.com/RorySullivan1/claudeBrain/tree/main/example-project/.claude/skills/session-memory
Command: npx skills add https://github.com/RorySullivan1/claudeBrain --skill session-memory-rorysullivan1

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

The skill provides a file-based memory system for Claude Code projects, enabling persistence of decisions, state, and open threads across sessions by using a dedicated .claude/memory/ directory and per-session logs.

Core Features & Use Cases

  • Auto-loaded INDEX.md that stores a concise, up-to-date memory state for the project.
  • Append-only session logs under .claude/memory/sessions/ capturing decisions, context, and next steps.
  • Quick recall of past work to continue conversations, reduce rework, and maintain continuity across sessions.

Quick Start

Initialize memory storage with python .claude/skills/session-memory/scripts/memory.py init and record a session with python .claude/skills/session-memory/scripts/memory.py new --slug <topic> --goal <goal>.

Frequently Asked Questions about session-memory

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

FAQPage Schema
How do I persist context and recall decisions across Claude Code sessions?

You can persist context across Claude Code sessions by using a file-based memory system that maintains a project state in an INDEX.md file and records detailed context in append-only session logs under a dedicated directory. This approach ensures past decisions are available when resuming work.

What is the best way to maintain session history for a coding project?

Maintaining session history is best achieved by logging decisions and next steps in append-only files within a dedicated memory directory. An auto-loaded index file provides a concise, up-to-date memory state, reducing rework and maintaining continuity across project sessions.

How do I initialize and record a new session log using a Python CLI?

To initialize session log storage, run the Python CLI memory script with the init command. Record a new session by executing the memory script with the new flag, providing a topic slug and a goal to capture decisions and context.

Do I need a specific directory structure to use file-based project memory?

Yes, file-based project memory requires a dedicated .claude/memory directory containing an INDEX.md file for state and a sessions/ subdirectory for logs. This structure is necessary for the Python CLI to properly initialize, log, search, and index session data.

How does indexing past work help when resuming tasks after a break?

Indexing past work helps when resuming tasks after a break by providing a concise, up-to-date memory state that allows for quick recall of past decisions and context. This reduces rework and maintains continuity across multiple project sessions.