AIKB — AI Knowledge Base

Shared context across sessions, tools, and machines.

---

Shared, persistent, inspectable memory for your AI tools.

AIKB gives your agents shared context that survives across sessions, tools, and machines. It stays local-first, Git-backed, and fully inspectable, so your memory system feels like infrastructure you control instead of a black box you hope is right.

Your AI tools should not start from zero.

AIKB is designed as a **cross-agent, cross-machine memory layer** — owned by the operator, not the platform.

[**Get Started**](#quick-start) • [**How It Works**](#how-it-works) • [**Tool Support**](#ai-tool-compatibility)

---

What AIKB Is

AIKB is shared memory for your AI tools. It helps Claude Code, Gemini CLI, Codex, OpenCode, Cursor, and similar agent-style tools keep context across sessions and machines without making you repeat yourself every time.

The memory lives in plain markdown files in a Git-backed repo you control, so it is easy to inspect, search, and fix when something goes wrong. One of AIKB's biggest advantages is that memory is editable with ordinary tools. If the agent learns something wrong, you don't need a proprietary console — just open the repo, find the bad memory, and correct it.

Why People Use It

Every AI session tends to start from zero. You re-explain your projects, your