OpenClaw

OpenClaw is a personal AI assistant you run on your own devices: one agent, reachable on WhatsApp, Telegram, Slack, Discord and more. With a multi-channel agent, memory tends to split per channel. The Korely plugin keeps it in one place: install it, select it as your memory backend, and the agent carries the same memories and facts into every conversation.

korely-openclaw goes into OpenClaw's memory slot, which holds one plugin at a time. Once selected, your agent gets six memory_* tools backed by Korely, plus automatic recall before each turn.

Requirements

You also need a Korely API key. Get one free: the Hobby plan costs nothing.

Install

Once the plugin is published (not yet, see above), one command installs it:

openclaw plugins install korely-openclaw

Select it as your memory backend

OpenClaw has one memory slot. Point it at korely and add your key (use ${KORELY_API_KEY} or a SecretRef, not the key in plain text), then restart the gateway:

{
  plugins: {
    slots: { memory: "korely" },
    entries: {
      korely: {
        enabled: true,
        config: {
          apiKey: "${KORELY_API_KEY}",
          userId: "default",  // the person the memories are about (see the table below)
          autoRecall: true,    // inject memory before each turn (default true)
          autoCapture: false,  // store the reply after each turn (default false)
          topK: 6,
        },
      },
    },
  },
}

openclaw gateway restart

What your agent gets

With Korely selected as the memory backend, the agent can:

• Use the memory_* tools: memory_search, memory_add, memory_get, memory_list, memory_update and memory_delete. They call Korely directly. Korely writes the facts and closes replaced ones on its own side, about ten seconds after a write (extraction).
• Auto-recall. Before each turn, Korely's get_context assembles the facts that are true now plus the most relevant memories, and the plugin adds them to the prompt as a <korely-memory> block (a context block). No language model runs on Korely's side when it reads.
• Auto-capture (off by default). When you turn it on, the agent's reply is stored as a memory after each successful turn, and Korely writes facts from it like from any other memory.

Example session: WhatsApp

Here is auto-recall in practice. You message your agent on WhatsApp. Before the model runs, the plugin adds the relevant context, and the agent answers from your own history. The text below is an example.

You › When is my lease renewal due, and what did the landlord
say about the rent?

# auto-recall: get_context runs before the model and injects a
# <korely-memory> block with the facts that are true now
← injected context (abridged, from its memories section):

   "Lease renewal call with Anna (2026-05-28): Renewal deadline is
   July 1. Anna confirmed a 3% increase, 1,200 to 1,236 EUR per
   month starting August. She wants the signed copy by post."

Agent › Your lease renewal is due July 1. On the May 28 call Anna
confirmed a 3% increase, 1,236 EUR per month starting in August,
and she asked for the signed copy by post. Want me to create a
task with a reminder for the last week of June?

Read-only vs read-write

An agent that anyone can message on several channels is exposed to more input than a coding assistant on your laptop. That is why autoCapture is off by default. The agent reads memory (auto-recall and memory_search) but writes only when it calls memory_add, memory_update or memory_delete. Every stored reply also counts as a write against your monthly quota. Once you trust the loop, turn on auto-capture to store replies automatically:

config: {
  apiKey: "${KORELY_API_KEY}",
  autoRecall: true,
  autoCapture: true,   // now replies are stored after each successful turn
}

Configuration

Troubleshooting

Why Korely

For the full method surface, see the SDK reference. For how facts, supersession and time work, see temporal facts.