# Claude-Mem: The #1 GitHub Trending Tool That Gives Claude Code Perfect Memory

**23,000+ stars in days. 2,600+ stars TODAY alone.**

Claude-Mem just exploded onto the scene as GitHub's #1 trending repository. Here's everything you need to know about this game-changing Claude Code plugin.

## What is Claude-Mem?

Claude-Mem is a persistent memory compression system for Claude Code. It automatically:

1. **Captures** everything Claude does during your coding sessions  
2. **Compresses** it with AI (using Claude's agent-sdk)  
3. **Injects** relevant context back into future sessions

The result? Claude actually _[remembers](/content/articles/ai-agent-memory-why-2026-is-the-year-of-persistent-context/index.html)_ what you worked on yesterday. No more re-explaining your project every session.

## Why Developers Are Going Crazy For It

### The Problem It Solves

Every Claude Code user knows the frustration: you spend an hour getting Claude up to speed on your codebase, only to have it forget everything when you start a new session.

Claude-Mem fixes this permanently.

### Key Features

- **🧠 Persistent Memory** - Context survives across sessions  
- **📊 Progressive Disclosure** - Layered memory retrieval with token cost visibility  
- **🔍 Skill-Based Search** - Query your project history with natural language  
- **🖥️ Web Viewer UI** - Real-time memory stream at localhost:37777  
- **🔒 Privacy Control** - Tag sensitive content to exclude from storage  
- **🤖 Automatic Operation** - No manual intervention required

## How It Works

Claude-Mem uses a sophisticated hook system:

1. **SessionStart** - Loads relevant context from previous sessions  
2. **UserPromptSubmit** - Tracks what you're asking  
3. **PostToolUse** - Captures Claude's actions and observations  
4. **Stop/SessionEnd** - Compresses and stores the session

All data is stored locally in SQLite with full-text search (FTS5) and Chroma vector database for intelligent semantic retrieval.

## Installation (30 Seconds)

In any Claude Code session:

```
/plugin marketplace add thedotmack/claude-mem
/plugin install claude-mem
```

Restart Claude Code. Done.

## The 3-Layer Search Workflow

Claude-Mem's MCP tools use a token-efficient approach:

1. **search** - Get compact index (~50-100 tokens/result)  
2. **timeline** - See chronological context around interesting results  
3. **get_observations** - Fetch full details only for filtered IDs

This results in ~10x token savings compared to fetching everything.

## Why This Matters

Claude-Mem represents a fundamental shift in how we interact with [AI coding assistants](/content/articles/cursor-ai-vs-windsurf-vs-claude-code-2026/index.html). Instead of treating each session as isolated, we can now build genuine continuity.

Imagine:

- Claude remembering that bug you fixed last week  
- Instant context about your architecture decisions  
- No more "let me look at the codebase again"

## The Numbers Don't Lie

- **23,264 stars** (and climbing fast)  
- **1,529 forks**  
- **2,638 stars gained TODAY**  
- **#1 on GitHub Trending**

This is the fastest-growing Claude Code plugin ever.

## Who's Behind It?

Created by Alex Newman (@thedotmack), Claude-Mem is open source under AGPL-3.0. The project has attracted contributors from around the world with translations in 20+ languages.

## Should You Use It?

**Yes, if you:**

- Use Claude Code regularly  
- Work on projects over multiple sessions  
- Hate re-explaining context  
- Want Claude to remember your coding patterns

**Maybe not if you:**

- Only use Claude for one-off tasks  
- Have strict data locality requirements (though it's all local)

## Getting Started

1. Install via the plugin marketplace  
2. Restart Claude Code  
3. Start coding - Claude-Mem handles the rest

Check out the full documentation at [docs.claude-mem.ai](https://docs.claude-mem.ai/)