LLM-Base-Prompts
A personal collection of LLM system prompts, agentic personas and prompt-engineering experiments, including a set of prompts for guiding image-to-video generation from an existing image using vision-capable models.
This repository collects system prompts and cognitive-framework experiments the author built primarily with Claude, stated to also work with other leading language models. One section, called ACE-HFS, describes a projected simulation hybrid command-line concept for creating, modifying and deleting fragmented cognition to reduce unwanted hallucinations, explicitly noting that such a projected filesystem does not persist and is limited by the model's attention window.
A separate section of prompts is intended for use with vision-capable language models, to guide the creation of video-generation prompts based on an existing image. The collection also links to the author's other, related projects covering narrative generation and MCP-based narrative enhancement.
It is written in TypeScript, and the project's license is not asserted in its repository metadata.