ozankasikci

Global Agent Memory — AI skill for Claude Code

AI community

Local-first, project-aware memory MCP server for Claude Code, Codex, and other AI agents, with Obsidian and an owner dashboard.

How to install Global Agent Memory

This entry records only its repository, not the path inside it, so there is no exact command to give. Open ozankasikci/global-agent-memory and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Global Agent Memory does

Local-first, project-aware memory MCP server for Claude Code, Codex, and other AI agents, with Obsidian and an owner dashboard.

Alternatives in AI

Xerj — Local-first search engine and MCP server for AI agents 12.5k ★
Investment News — 为 A股投资者打造的全球产业链资讯看板 · 12 大赛道一一对应 A股板块(半导体/AI/机器人/新能源车…)，覆盖 100+ 权威源，用你自己的大模型每日提炼中文「今日要点」+ 翻译 · 全程本地、零 API key 447 ★
LumaBrowser — Local AI desktop app: a Chromium browser with a GPU-aware llama.cpp runtime, OpenAI-compatible API, AI chat an 125 ★

README

Global Agent Memory

Local-first, project-aware, durable memory for Claude Code, Codex, and other MCP-compatible agents.

[![Python 3.12+](https://img.shields.io/badge/Python-3.12%2B-3776AB?style=flat-square&logo=python&logoColor=white)](https://www.python.org/downloads/) [![PyPI](https://img.shields.io/pypi/v/global-memory-mcp?style=flat-square&logo=pypi&logoColor=white)](https://pypi.org/project/global-memory-mcp/) [![MCP contract](https://img.shields.io/badge/MCP-contract%20v1-5A67D8?style=flat-square)](docs/mcp-contract-v1.md) [![Local first](https://img.shields.io/badge/storage-local--first-2F855A?style=flat-square)](#security-model) [![License](https://img.shields.io/badge/license-MIT-blue?style=flat-square)](pyproject.toml) [![GitHub release](https://img.shields.io/github/v/release/ozankasikci/global-agent-memory?style=flat-square)](https://github.com/ozankasikci/global-agent-memory/releases/latest)

Global Agent Memory gives multiple coding agents one shared, reviewable memory without handing control of your knowledge base to a hosted service. Markdown files are canonical, a hybrid local MCP runtime is the public agent interface, and the dashboard lets a human approve, edit, protect, or remove what agents remember. Agent commands work without a background server. The authenticated localhost server starts on demand for the dashboard and can optionally run continuously for live file watching and shared embedding work.

**One Vault. Multiple agents. Human-controlled memory.**

![Global Agent Memory connects AI agents through a daemonless-capable hybrid MCP runtime to owner-controlled Markdown and dashboard surfaces](docs/assets/global-agent-memory-flow.svg)

[Quick start](#quick-start) · [Obsidian Vault](#obsidian-vault) · [Dashboard](#dashboard) · [How it works](#how-it-works) · [Security](#security-model) · [MCP contract](#mcp-interface) · [Documentation](