Bozo AIGC Prompt Enhancer

Problem

A lot of unstable AIGC output is not caused by the model being weak, but by under-specified prompts: one goal, but no task scope, output format, constraints, examples, or acceptance criteria. This kind of prompt forces the model to guess, which can produce off-topic answers, over-generalized content, inconsistent length, or results that are hard to reuse. The issue becomes more visible when the request comes from a verbal brief, meeting notes, or a one-liner like “help me write this.” bozo-prompt-enhancer targets this input-side problem by turning brief requirements into clearer model instructions.

How It Works

It usually works around the original prompt, structuring and checking it:
- identifies the task type, audience, and final output form
- adds tone, length limits, prohibitions, and delivery criteria
- checks for missing boundaries, role constraints, or unrelated context

For vague input such as “write a plan,” it can expand the request into an instruction with goal, audience, section structure, reasoning basis, and deliverable format. For an already detailed prompt, it can help surface gaps, such as missing success criteria, blurred facts versus assumptions, or unclear output order. In practice, you can start with a one-sentence goal, ask for a paste-ready prompt version, and then trim redundancy based on model feedback.

Boundaries

The skill is best used as a prompt-drafting step before calling a model, not as a substitute for domain expertise. If the original requirement lacks key parameters, the enhanced version may contain reasonable assumptions and should be reviewed manually before production use. It is not a fixed template library; it is a way to organize and clarify natural-language instructions. For high-risk domains such as compliance, healthcare, finance, or legal work, professional standards and verified data still take priority.