Vertical-drama scriptwriting toolkit for AI — teardown templates, hook formulas and dialogue prompts that turn binge-drama patterns into sellable episode scripts
LLMs write scripts that read like scripts. Vertical dramas (60-90 second episodes, 60-100 per series) don't pay for prose — they pay for patterns: a hook in the last 10 seconds of every episode, a reversal every 3 episodes, an identity reveal every 5, and dialogue lines short enough to be screenshotted. This toolkit encodes those commercial patterns into five prompt templates and five working techniques, so the model writes to the format instead of fighting it: - Binge-drama teardown template — dissect 10 hit shows into a reusable pattern sheet before writing a word - Episode structure prompt — locks each episode's goal, conflict and closing hook before generation - Character bio prompt — builds flawed, screenshot-able leads (a defect, a habit, a relational weakness) instead of perfect nobodies - Dialogue polish prompt — rewrites "correct" lines into short-drama lines: under 15 characters, one emotional beat, quotable - Client delivery template — requirement-confirmation sheet that prevents the "feels off" revision loop Plus five pro techniques with ready prompts: 3-episode mini-climax / 5-episode major-climax pacing, information-gap suspense, the hook formula (unexpected event + unfinished action + open question), and the three-pass revision loop (structure, then dialogue, then hooks). Ships as an Agent Skill (SKILL.md) so Claude Code / Cursor can run the whole loop, or as plain markdown templates. MIT licensed.
$ git clone https://github.com/nianjiang888/ai-scriptwriting-skill.gitReads all knowledge files and MCP integrations to produce a single-page product status dashboard with OKR progress, sprint status, decisions, blockers, launches, and metrics.
Research a competitor from their URL or name. Fetch website, pricing, features, and news. Generate a battlecard with positioning, strengths, weaknesses, and strategic implications.
Generate meeting preparation docs tailored by meeting type, pulling context from past meetings, recent decisions, action items, and sprint status.