Use this skill when an operator wants Claude Code to retain useful project context across sessions. The agent installs and configures Claude Mem, verifies that coding-session activity is being captured, lets Claude Mem summarize or compress past work, and checks that relevant memory is injected into later Claude Code sessions before planning or editing.

## When to invoke

Invoke this instead of relying on normal Claude Code chat history when the work spans multiple sessions, branches, or recurring project routines and the agent keeps re-learning decisions, file structure, conventions, or unfinished tasks. It is best for coding workflows where continuity and compressed recall reduce repeated exploration.

## Scope boundary

This is not a generic memory-product listing. The skill is bounded to Claude Code session capture and retrieval: install, configure, verify capture, review compressed memories, and use injected context during future coding-agent work. Do not use it as a broad knowledge-base, CRM, or multi-agent memory architecture without an explicit Claude Code session workflow.

## Source-backed signals

Source: https://github.com/thedotmack/claude-mem. Upstream describes Claude Mem as a persistent memory compression system built for Claude Code that captures session activity, compresses it with AI, and injects relevant context into future sessions. The verified repository has 74,413 GitHub stars, recent push and release activity, Apache-2.0 license, and documentation at https://claude-mem.ai.

## Installation

Follow the upstream installation instructions at https://claude-mem.ai and the repository README. Confirm the Claude Code integration is active before depending on memory injection for production-impacting edits.