What is the Apiary?

The plain-language introduction to the whole Apiary suite. Start here if you want the big picture before reading about any one product.

What is the Apiary?

The plain-language introduction to the whole Apiary suite. Start here if you want the big picture before reading about any one product.

Related:


#The one-line version

The Apiary is a stack of small, sharp programs that give your AI coding agents one shared, lasting memory on hardware you control.

#The problem it solves

AI coding assistants are brilliant in the moment and forgetful the next. Close the window and the context is gone. Open a different tool tomorrow and it has never heard of your project. So you re-explain your conventions, re-discover fixes that already worked, and repeat yourself to a machine that should have remembered. The Apiary ends the forgetting.

#Five products, one memory

Each product does one job well, and together they work as one system.

  • Honeycomb is the memory. It quietly records what happens as you work, distills it into clean notes, and hands them back to any assistant that asks.
  • Nectar is the understanding. It gives every file a plain-language description so your agent can find code by meaning, not by file name.
  • Hive is the one portal you open. A single dashboard for the whole stack at one address.
  • Doctor is the watchdog. It keeps the local daemons alive and heals what breaks, and only speaks up when it genuinely cannot fix something.
  • Queen is the cloud orchestrator for when the stack outgrows a single machine. It sees and steers a fleet without ever reading your memory.

#What makes it different

  • It runs on your machine. The daemons are local; your memory lives in a store you own.
  • It installs in one command. No account to create first, no server to host, no database to run.
  • It works underneath the tools you already use. Claude Code, Cursor, and Codex today, with Hermes, pi, and OpenClaw in progress.
  • It gets sharper over time. Memory is tidied, lessons become shareable skills, and your codebase becomes searchable by meaning.
  • It is honest about your data. Almost nothing leaves your machine, and what does is scrubbed and optional.

#Who it is for

  • Individual developers who want their assistant to remember their project.
  • Teams and enterprises that want many developers, devices, and tools to learn as one brain, with data separated cleanly by team and project.

#Where it comes from

A collaboration between Legion Code and Activeloop. Activeloop provides Deep Lake, the database for AI the memory lives in, and Hivemind, the open-source agent-memory project Honeycomb builds on. Legion Code adds the multi-tier memory, the skill sharing, the self-tidying loop, and the daemons that tie it all together.

#Next steps