Big Prompt Hub – Sharing AI Prompts
Static prompt directory, but every major AI tool has a free prompt library now.

They combined a big prompt library (500+ Grok prompts) with a simple one-click generate flow and a scene-by-scene ‘extend’ pipeline for building short films — a UX that actually maps to how creators iterate. The landing copy promises fast, 4K-aware renders and multi-input support, but the pitch feels derivative in a crowded market and key details (model provenance, export limits, real sample output quality) are missing.
AI artists, indie filmmakers, social media creators, designers and hobbyist content creators
Static prompt directory, but every major AI tool has a free prompt library now.
Fast, mobile-first interface that actually uses Zigbee groups and scenes to make lights change together instead of the staggered multi-device flicker most UIs produce. Automations run locally inside Zigbee2MQTT and the PWA caches state for instant launches — thoughtful touches like per-group permissions and client-side password hashing show the author worked through real deployment pain. Remote access via UPnP is a neat convenience, but it’s also the one place you should audit before exposing your network.
Wraps Grok image-to-video for sprites, but Aseprite and Leonardo already dominate this niche.
The pitch — scene-by-scene intensity maps benchmarked against 100+ influential films — is a tidy, useful idea for writers who want empirical beat sheets. The landing is visually strong and sells the concept, but the web app currently fails to load (client-side exception) and there's no visible explanation of how 'intensity' is measured or scenes are detected. Fix the runtime reliability and publish methodology/dataset and this could move from intriguing demo to genuinely useful niche tool.
npm for AI agent prompts with commit-pinned lockfiles, but still early and experimental.
You get momentum scores, RSI, EMA alignment, coil-breakout detection and concise bull/bear briefs out of the box — plus an OpenClaw skill so an LLM agent can answer “how's $NVDA looking?” immediately. The author ships 900+ days of backtested signals and open, dependency-free Python scripts which is refreshingly transparent, but the headline win-rate/return claims need independent audit (survivorship and lookahead bias are the obvious caveats) and the product looks focused on US/end-of-day use-cases.