A sufficiently decentralized social network. farcaster.xyz

we now have miniapps.farcaster.xyz/llms-full.txt and `pnpm create @farcaster/mini-app`

what's the best way to integrate these two? when we setup a starter repo should we have a claude.md? what about other agents? what should go in that setup file?

Building @warpcast and @farcaster / dad / like to read, cook, ski, code

Developer ecosystem @farcaster, co-founder @bountycaster. Previously: co-founder of Scalar Capital, product at Coinbase, co-producer @ethereumfilm

Decentralized open source consciousness | Values-aligned communities | citizen of Black Rock City

Already gave this feedback to @linda but the hello world version of this didn’t work for me, but the Neynar one did

I would prefer to keep it generic for all LLMs, instead of cursorrules or claude.md.

I usually use a copy from docs in farcaster.md and neynar.md files.

dev + founder | @vibesengineering.eth prev: @onsenbot @herocast

can you say more? or even better what specific changes if any would you make to this template repo (i.e. changes to readme, new MD files) to make it more vibe codable from the start?
https://github.com/farcasterxyz/miniapps/tree/main/packages/create-mini-app/templates/default

@hellno.eth can probably offer advice from building vibes.engineering

cursor rules please!

although llms-full.txt should suffice, cursor rules are super helpful

do not build for specific llms.. i dont really know what you’re asking outside that

This is clean. are we assuming most devs will want to integrate a single LLM, or would a multi-agent setup be the move? Maybe a modular llms/ directory instead of one markdown file?