🦞 OpenClaw

🦞 OpenClaw#

Warning

This integration is still under active development and may change without notice.

For your comments and suggestions, contact us by email at bot@olvid.io, or visit our GitHub repository: openclaw-channel-olvid.

Installation#

If you already have a functional OpenClaw instance, all you need to do is install and configure a daemon, then install and configure our OpenClaw plugin. Otherwise, refer to the docs for installing OpenClaw with Docker.

Danger

Beware when installing and using OpenClaw (and any AI capable of executing commands and code). Here are our minimum security recommendations:

  • install OpenClaw in a Docker container.

  • use only locally hosted AI models (ollama or other).

  • use the Olvid channel to secure your exchanges with your agent.

  • do not use skills vulnerable to “prompt injections” (i.e., those that use data defined by a third party, such as an email).

Daemon#

Here is a docker-compose.yaml file to install a local Olvid daemon. If OpenClaw is installed with Docker, simply add the daemon and cli services to your existing docker-compose.yml file.

services:
  daemon:
    image: olvid/bot-daemon:2.0.1
    environment:
      - OLVID_ADMIN_CLIENT_KEY_CLI=SetARandomValue
    ports:
      - 50051:50051
    volumes:
      - ./daemon-data:/daemon/data
  cli:
    image: olvid/bot-python-runner:2.0.1
    entrypoint: "olvid-cli"
    environment:
      - OLVID_ADMIN_CLIENT_KEY=SetARandomValue
      - OLVID_DAEMON_URL=http://daemon:50051
    stdin_open: true
    tty: true
    profiles: ["cli"]

Let’s now launch our Olvid daemon.

docker compose up -d daemon

We can then launch the olvid CLI in interactive mode.

docker compose run --rm cli

You can then create your bot with the following command, following the on-screen procedure.

identity new Claw Bot

To finish and quit the Olvid CLI, use the exit command or press ctrl+d.

Olvid Channel#

One can then move on to the installation and configuration of the Olvid plugin in OpenClaw.

openclaw plugins install olvid-channel
openclaw plugins enable olvid-channel
openclaw channels add --channel olvid --url DAEMON_URL --token BOT_TOKEN

⚠️ You need to replace DAEMON_URL with http://daemon:50051 (or http://localhost:50051 for a local installation) and BOT_TOKEN with the previously created client key.

That’s great! If everything went well, the messages you send to your Bot in Olvid are forwarded to your OpenClaw agent. 🥳

Note

If you have lost your client key, use the following command.

docker compose run --rm cli key get -f key

Securing your installation#

Section under construction

Configuration of OpenClaw#

If your OpenClaw instance is used by multiple channels or users, use the per-account-channel option for the session.dmScope field. This allows you to create a session per Olvid conversation and avoid mixing contexts.

https://docs.openclaw.ai/gateway/security/index#secure-dm-mode-recommended