Everyday use

Once the apiary is set up, the day-to-day experience is mostly about not having to think about it. This page covers what actually happens while you work.

#What do I have to do day to day?

Almost nothing. Once your assistants are wired with honeycomb setup, the apiary captures the useful moments as you work and hands the right notes back at the start of your next session. You do not have to remember to save anything for the basics to keep working.

What you will notice is a fresh session starting already informed, your assistant already knows the project's recent decisions and durable conventions instead of asking you to re-explain them.

#What does a session start with, once things are set up?

Your assistant opens with a short briefing, the prime, of what the apiary already knows about this project. It is deliberately small, just headlines pulled from the top tier of memory, so it informs without cluttering the conversation. If your assistant needs more than the headline, it asks for the fuller summary or the raw detail on demand.

#How do I save something on purpose?

Sometimes you want to pin a fact down yourself instead of waiting for capture to notice it. Two commands do the whole job:

honeycomb remember "the staging database resets every night at 2am UTC"
honeycomb recall "staging database schedule"

remember writes a clean note. recall pulls back whatever is relevant to your question, and it does not need to match the exact wording of the note you saved. Once meaning-search has warmed up, recall matches on meaning too, so you can find something even when you would not have guessed its exact phrasing.

#What does the hive dashboard show me?

Open it any time with honeycomb dashboard, or just keep the tab open. It is one page for the whole picture:

• home for the at-a-glance view: how much has been remembered, and overall health.
• harnesses for each connected assistant and whether its wiring is healthy.
• memories for your captured knowledge, browsable.
• graph for the map of your codebase.
• sync for the skills and assets being shared and pulled.
• logs for a live view of what the daemons are doing.
• settings for preferences and sign-in status.

If anything looks off, the dashboard says so plainly rather than showing a green light that is not telling the truth.

#What are skills, and do I have to manage them?

When you solve something genuinely reusable, a migration trick, a debugging routine, the apiary can capture it as a skill: a short, reusable lesson. Skills you or your teammates create show up automatically in your assistants at the start of a session, placed wherever each tool looks for them. You do not have to copy files around.

You can manage skills directly if you want more control:

honeycomb skill pull
honeycomb skill scope team --users alice,bob

#Does it work the same across different tools and machines?

Yes. Every assistant talks to the same local daemon and the same store, so a memory written from one tool is recalled by another, and a memory captured on one machine is available on another, as long as you are signed in with the same account. Switch from one assistant to a different one mid-project and the context comes with you.

#Can I turn things up or down?

Two capabilities are off by default so nothing surprising happens on day one. Semantic search, finding memories by meaning, turns on by itself once its small local model has downloaded and warmed up. Until then, recall still works by matching words. The self-tidying loop, which merges duplicates and prunes stale notes over time, is opt-in because it uses a model, so you decide when to spend that.

If you are working on something sensitive, you can put the apiary into read-only mode for a session. Recall still works, but nothing new is written, and your assistant keeps working normally either way.

#Common questions

#What do I have to do day to day?

Almost nothing. Once your assistants are wired, the apiary captures useful moments as you work and hands the right notes back automatically at the start of your next session. You only reach for a command when you want to pin something down on purpose.

#What does a session start with, once things are set up?

Your assistant opens with a short briefing of what the apiary already knows about the project, its recent decisions and durable conventions, instead of starting blank. It is deliberately small, just headlines, and the assistant can ask for more detail whenever it needs it.

#How do I save something on purpose?

Run honeycomb remember followed by the fact, then honeycomb recall followed by a question to get it back. This is for things you want pinned down deliberately, on top of what gets captured automatically.