LP: Extend OpenClaw for a Local-First AI Assistant

[odincodeshen]

Before submitting a pull request for a new Learning Path, please review Create a Learning Path

• [ x ] I have reviewed Create a Learning Path

Please do not include any confidential information in your contribution. This includes confidential microarchitecture details and unannounced product information.

• [ x ] I have checked my contribution for confidential information

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of the Creative Commons Attribution 4.0 International License.

[kieranhejmadi01]

I've validated the steps below to configure telegram with the DGX Spark on my personal telegram account.

---

Create a Telegram account if you do not already have one.

Create a new Telegram bot
* Open Telegram on a client device, telegram is available on a variety of platforms.
* Start a chat with BotFather.
* Send:
/newbot
* Follow the prompts to create your bot.
* Once the bot is created, copy the HTTP API token provided by BotFather. You will need it in the next steps.

From your Telegram client, open a chat with the bot you just created and send it a test message, such as Hello. This creates an update that you can use to retrieve the bot’s CHAT_ID. Next, open a terminal on your local machine and run the following command, replacing TOKEN with the HTTP API token provided by BotFather:

curl https://api.telegram.org/bot<TOKEN>/getUpdates"

you should see an output like the following, copy the id value.

{
    "ok": true,
    "result": [
{
            "update_id": 78772718,
            "message": {
                "message_id": 2,
                "from": {
                    "id": 8974517049,
                    "is_bot": false,
                    "first_name": "Kie",
                    "last_name": "Hej",
                    "language_code": "en"
                },
                "chat": {
                    "id": 8974517049,
                    "first_name": "Kie",
                    "last_name": "Hej",
                    "type": "private"
                },
                "date": 1785762058,
                "text": "Hello?"
            }
        }
]
}

Use the id value and set as your-telegram-chat-id in the .env file

[kieranhejmadi01]

The first time you run docker compose up, vLLM downloads the ~30 GiB Qwen model, so startup can take a significant amount of time before the runtime becomes available. The reported “1–3 minute” startup time is therefore misleading on the initial run. Should let the reader know that is can take a while for the vLLM service to be available