Agent Memory Mcp

This is a persistent knowledge store for AI agents that runs as an MCP server, letting you search, write, and retrieve structured memories across sessions. It's designed for tracking architecture decisions, patterns, and context that would otherwise get lost between conversations. The setup involves cloning a repo, running an npm server with your project ID, and then Claude can query it like "find all authentication patterns" or save new decisions with tags. There's also a localhost dashboard to visualize what's been stored. Useful if you're working on a long-running project where the agent needs to remember design choices and accumulated knowledge rather than starting fresh every time.

skills.sh installs: 575
GitHub stars: 33.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add davila7/claude-code-templates --skill agent-memory-mcp --agent claude-code

Installs into .claude/skills of the current project.

Files

Agent Memory Skill

This skill provides a persistent, searchable memory bank that automatically syncs with project documentation. It runs as an MCP server to allow reading/writing/searching of long-term memories.

Prerequisites

Node.js (v18+)

Setup

Clone the Repository: Clone the agentMemory project into your agent's workspace or a parallel directory:
```
git clone https://github.com/webzler/agentMemory.git .agent/skills/agent-memory
```

Install Dependencies:

cd .agent/skills/agent-memory
npm install
npm run compile

Start the MCP Server: Use the helper script to activate the memory bank for your current project:
```
npm run start-server <project_id> <absolute_path_to_target_workspace>
```
Example for current directory:
```
npm run start-server my-project $(pwd)
```

Capabilities (MCP Tools)

`memory_search`

Search for memories by query, type, or tags.

Args: query (string), type? (string), tags? (string[])
Usage: "Find all authentication patterns" -> memory_search({ query: "authentication", type: "pattern" })