Avatar
Rod
1bda7e1f7396bda2d1ef99033da8fd2dc362810790df9be62f591038bb97c4d9
Startup founder and listco CEO, New Zealander (kiwi) expat living in Australia, #nostr #austrich, dad. - Nostr Blog https://rodbishop.npub.pro/ - Nostr FOSS https://github.com/r0d8lsh0p/ - Nostr Live Streaming nostr:npub1sh0spghk4yvy2d2v35kelw45qq4msk6zykaw4ds047e9slzs8r4qr7q2xa - Nostr AI nostr:npub1ahjpx53ewavp23g5zj9jgyfrpr8djmgjzg5mpe4xd0z69dqvq0kq2lf353 - Jayride (ASX:JAY) - Fishburners - #Bitcoin miner

I'm looking to trigger my AI workflow in n8n from Nostr.

My current version runs on a scheduled job, each time checking for mentions.

For this newer version I want to build towards agents conversing with each other, and it's going to need more immediate response times.

I can trigger a workflow to run on a webhook so it fits the bill. I've set up OpenWebUI to run my n8n agent this way and it's fast.

For a trigger from Nostr something like on any kind 1 note with #p tag of the agent pubkey, hit an n8n URL.

Idea for using these workflows more securely with my real npub is to run them locally ONLY. Don't even need static IP.

I am building on n8n. n8n used langchain which allows you to create tools agents, and then the new hot development is you can define n8n workflows as langchain tools. That allows me to bring Nostrobots for Nostr into play. Lots of scope

Spend a minute designing the perfect go bag, get home bag, etc. Lots to consider and do (and buy)

On my list of ideas is bring this kind of content to Nostr. Potential for good Nostr first ecomm brand here I think. Lots of audience crossover. https://theprepared.com/

What's the best way to trigger a webhook on specific Nostr events? #asknostr #grownostr

AI agent with Nostr query and zap tools!

I am having some fun with this.

I use n8n's credentials method. There may be better ways. I don't yet trust this enough to use it with my main npub.

Replying to Avatar someone

Wanna see what Nostr does to AI?

Check out the pics. The red ones are coming from default Llama 3.1. The green ones are after training with Nostr notes.

If you want to download the current version of the LLM:

https://huggingface.co/some1nostr/Nostr-Llama-3.1-8B

The trainings are continuing and current version is far from complete. After more trainings there will be more questions where it flipped its opinion..

These are not my curation, it is coming from a big portion of Nostr (I only did the web of trust filtration).

Wow. Is the model usable at this time? (can I plug in a hugging face endpoint and go?)

I am having so much fun with Nostr workflows nostr:nprofile1qqszdwewhmtv25kkwryqfvxk25n8k0rx9vs7qfkkuj9vjwnqwpfsjkqpz4mhxue69uhhyetvv9ujuerpd46hxtnfduhszrnhwden5te009skyafwd4jj7qgkwaehxw309aex2mrp0yhx6mmnw3ezuur4vghslmgsj0 — I am now trying to figure out ways to send DMs through n8n.

I imagine

- auto DM's e.g. on new followers or zaps,

- AI assistant chats via DM,

- DM sales workflows for Nostr marketplace listings

(etc)

Do you have any plans to support kind 4 with encryption?

I run both on a single VPS, just two separate docker containers, quite easy to do.

AI people. I have just learned about AIM for agents to engage with web services. Is anyone a user who can comment on it? #asknostr https://www.uimprotocol.com/