agentMemory

Synchronize markdown memory banks across KiloCode, Cline, and RooCode via MCP.

30|5|Updated Dec 28, 2025
One-click install
npx skills add https://github.com/webzler/agentMemory --skill agentmemory-webzler
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agentMemory
Source: https://github.com/webzler/agentMemory/tree/main
Command: npx skills add https://github.com/webzler/agentMemory --skill agentmemory-webzler

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI coding agents often lose context across sessions, leading to hallucinations and duplicated effort. agentMemory provides a centralized, persistent memory store that stays in sync with project documentation and is fully searchable.

Core Features & Use Cases

  • Bi-directional synchronization between KiloCode, Cline, RooCode memory banks and the MCP-backed storage.
  • Fast, tag-based search and cross-project memory querying with analytics.
  • Visual dashboard and programmatic MCP tools for memory management and automation.

Quick Start

Install and initialize agentMemory in your workspace, start the MCP server, and begin storing and querying memories.

Frequently Asked Questions about agentMemory

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

FAQPage Schema
How do I add persistent memory to my AI coding agent to stop context loss?

Persistent memory for AI coding agents prevents context loss by providing a centralized store that stays in sync with markdown memory banks. This enables cross-session memory retrieval, eliminating duplicated effort and hallucinations across project sessions.

Can I sync markdown memory banks across KiloCode, Cline, and RooCode projects?

Yes, bi-directional synchronization connects KiloCode, Cline, and RooCode markdown memory banks with an MCP-backed storage. This allows you to manage and sync markdown memory files automatically across multiple coding agent environments.

How do I search cross-project memory tags for AI coding agents?

You can perform fast, tag-based search and cross-project memory querying using the memory_search MCP tool. The system also provides memory_stats for analytics, allowing you to retrieve specific context efficiently across all stored agent memories.

Does the agentMemory MCP server require an internet connection to manage memory?

No, the local MCP tooling operates offline-first, meaning memory_write, memory_read, and memory_update function without an internet connection. It includes robust error handling to ensure memory management continues reliably during network disruptions.

How do I initialize an MCP-backed memory store in my VS Code workspace?

To initialize the memory store, install agentMemory in your workspace, start the MCP server, and begin using the provided programmatic tools. You can then access the visual dashboard or use MCP commands directly to manage your agent's persistent context.