What is Aplexica?
A short introduction to Aplexica, the agent-agnostic memory layer for developers running more than one AI agent.
- By
- Aplexica
- Published
- Tagged
- product intro
If you have used an AI agent for more than a few months, you have a problem: the agent knows things about you, your codebase, and the way you like to work — and none of that is portable. Switch agents and you start over. Add a second agent and you maintain context in two places. Try a new agent the day it ships and the comparison is rigged before you begin.
Aplexica fixes this by making your agent state portable.
What it does
Aplexica is a small, local-first daemon you install on your machine. It watches the files each AI agent uses to store its memories, skills, and conversation history. When you change one, it propagates the change to the others. When you add a new agent, your full context is already there.
Three things make this work:
- Filesystem-native architecture. Aplexica writes into each agent’s native files. There is no central server holding your state hostage. Uninstall Aplexica tomorrow and every agent retains its full state.
- First-class adapters. Claude Code, Codex, Hermes, OpenClaw, Kilo — round-trip support for memories, skills, and conversation history. New agents added via the same adapter API.
- Forking workflows. Branch a conversation across two agents. See which answer you prefer. Continue with that branch.
Two editions
The Open Source edition is free forever. It runs entirely on your machine; the daemon makes zero network calls in its default configuration.
Aplexica Cloud adds optional zero-knowledge sync between devices, team namespaces, governance, and SSO. We can’t read your content even when you turn Cloud on — the keys never leave your devices.
Try it
The Open Source edition is free and installs from supported package managers; keep using whichever agents you already prefer. See the product page for the full feature surface, or jump straight to GitHub.