What is Agent Memory MCP?

It is a specialized capability that allows Claude to store and retrieve information about a project, such as design patterns and technical decisions, across different sessions.

Can I use it for multiple projects?

Yes, you can start the server with unique project IDs and absolute paths to maintain separate, isolated memory banks for different repositories.

Does it require an external database?

It runs locally as an MCP server using Node.js, storing the memory bank within your designated workspace or project directory.

How does the memory search work?

The skill provides a memory_search tool that allows users or the AI to find specific records using text queries, specific types (like 'decision'), or custom tags.

What is the purpose of the dashboard?

The dashboard provides a visual interface at localhost:3333 to monitor memory usage, view statistics, and manage the knowledge bank outside of the chat interface.

Agent Memory MCP

Name: Agent Memory MCP
Author: claudiodearaujo

byclaudiodearaujo

•

生产力与工作流

Manages a persistent, searchable knowledge bank for AI agents to track project architecture, patterns, and design decisions.

Agent Memory MCP provides a robust, long-term memory layer for AI agents, enabling them to store and retrieve contextual information such as architectural decisions, implementation patterns, and project-specific knowledge. By integrating as an MCP server, it allows Claude to maintain continuity across coding sessions, ensuring that past insights are instantly searchable and accessible. The skill includes a visual dashboard for memory analytics and simplified tools for reading, writing, and searching the project's collective intelligence, effectively acting as an external brain for your AI-driven development workflow.

主要功能

01Semantic search for memories by query, type, or tags

021 GitHub stars

03Automated synchronization with project documentation

04Standardized tools for recording architectural decisions and patterns

05Visual dashboard for memory usage and analytics

06Persistent long-term memory storage via MCP protocol

使用场景

01Tracking technical debt and implementation patterns to avoid redundant problem-solving

02Onboarding AI agents to existing projects by providing a searchable record of past decisions

03Maintaining architectural consistency across complex, multi-session development tasks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add claudiodearaujo/sistema-de-narra-o-de-livro agent-memory-mcp

For use in Claude.ai and ChatGPT

Download Skill