Getting started with honeycomb

Getting honeycomb running takes one command, one click in a browser tab, and one pair of commands to prove it works. No database setup, no config files to hand-edit.

#How do I install honeycomb?

Open a terminal and paste the line for your system.

macOS or Linux:

curl -fsSL https://get.theapiary.sh | sh

Windows (PowerShell):

irm https://get.theapiary.sh/install.ps1 | iex

You don't need Node or npm already set up. The installer detects what's missing and installs it. When it finishes, it opens the hive dashboard in your browser automatically. The terminal is just the doorway; the dashboard is where setup actually happens.

Want to read the script before you run it? Visit get.theapiary.sh in a browser first. It serves an inspect-before-piping page and publishes a checksum file so you can verify what you're about to run.

#Where do I finish setup?

In the hive dashboard, not the terminal. The dashboard runs locally at 127.0.0.1:3853, and it's the real starting point for honeycomb.

Once it's open, click First time setup. honeycomb shows a short code right on the page and opens a tab where you approve it (create a free Deeplake account here too, if you don't have one yet). You never copy a code out of a terminal window.

The moment you approve, the same dashboard tab lights up with your connected views. Nothing to restart, no second tab to chase down.

If you already have Hivemind installed, the dashboard notices and offers to move you over cleanly. Running both at once isn't supported, so let honeycomb handle the switch. It backs up your Hivemind setup first, and because the two share one sign-in, you usually won't need to sign in again.

#How do I connect my coding assistants?

From the hive dashboard or your terminal, run:

honeycomb setup

This finds the AI coding assistants you already have installed, Claude Code, Cursor, and Codex today, with Hermes, pi, and OpenClaw in progress, and wires each one so honeycomb captures and recalls automatically as you work. It's safe to run again any time, for example right after you install a new assistant; a re-run only touches what actually changed.

To wire just one assistant by name:

honeycomb connect claude-code

Check that everything connected:

honeycomb status

#How do I save my first memory?

Now teach honeycomb something and ask for it back:

honeycomb remember "we deploy from the release branch, never from main"
honeycomb recall "how do we deploy"

remember writes a clean note. recall pulls back whatever's relevant to your question, even if you phrase it differently than the original note. Write it while using one assistant, and a different assistant will recall it tomorrow, on another machine, if you're signed in to the same account.

#What happens after that?

Open a fresh session in one of your wired assistants, in the same project. Instead of starting blank, it begins with a short briefing of what honeycomb already knows: your recent decisions and durable conventions, including the note you just saved. You didn't have to ask for it.

From here, honeycomb mostly runs itself. Browse back to the dashboard any time with honeycomb dashboard to see your memories, the state of each connected assistant, your team's shared skills, a map of your codebase, and overall system health, all on one local page.

#Common questions

How do I install honeycomb? Run curl -fsSL https://get.theapiary.sh | sh on macOS or Linux, or irm https://get.theapiary.sh/install.ps1 | iex in Windows PowerShell. The installer handles Node and npm for you and opens the hive dashboard when it's done.

Where do I finish setup? In the hive dashboard at 127.0.0.1:3853, not the terminal. Click First time setup and approve the code shown on the page.

How do I save and recall my first memory? Run honeycomb remember "your note" to save it, then honeycomb recall "a related question" to get it back.