Hey!
I've been exploring using local crawling with local FireCrawl as an alternative to the Twitter API.
I'm then using a local LLM (meta-llama-3.1-8b-instruct in LMStudio) to generate drafts.
I've been testing the workflow in my fork here: fork, and it's been pretty smooth.
I think this approach could be beneficial and would love to discuss the possibility of adding this feature to the main repo!
(i've added tests too - i'll do a proper branch if that sounds interesting)
Hey!
I've been exploring using local crawling with local FireCrawl as an alternative to the Twitter API.
I'm then using a local LLM (meta-llama-3.1-8b-instruct in LMStudio) to generate drafts.
I've been testing the workflow in my fork here: fork, and it's been pretty smooth.
I think this approach could be beneficial and would love to discuss the possibility of adding this feature to the main repo!
(i've added tests too - i'll do a proper branch if that sounds interesting)