out-of-ideas — All AI Prompts for Review
# out-of-ideas — All AI Prompts for Review
This file collects every AI-facing prompt in the pipeline so an external reviewer can audit them for optimization potential. For each prompt: what it does, where it sits in the pipeline, why we wrote it that way, known production issues, and the prompt content itself.
---
## Project context (read this first)
**The product domain.** The pipeline targets a competitive 1v1 mobile trivia game profile — TikTok-native audience aged 14–30, ~10 seconds per question on small phone screens. The core thesis: **fun > educational > factual**. A correct-but-boring question is worse than no question.
**The pipeline.** A standalone Node CLI (TypeScript) that produces, scores, fact-checks, embeds, dedups, and ships quiz questions into a default-repertoire pool (`data/finalized-pool.jsonl`, currently ~465 entries). Six stages chain together; AI prompts gate every stage that uses an LLM:when to use it
Community prompt sourced from the open-source GitHub repo EmilHerzberg/out-of-ideas (Apache-2.0). A "out-of-ideas — All AI Prompts for Review" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.
tags
roleplaycommunitygeneral
source
EmilHerzberg/out-of-ideas · Apache-2.0