Markdown Output Prompt — Lock the Document Structure
The contract that stops AI documents from restructuring themselves: a pinned section skeleton, forced tables, and strict consistency rules.
Write trust, social proof, risk reversal, and objection-handling copy for a landing page — with proof placeholders and an explicit do-not-invent rule, so nothing is fabricated.
Social proof is the most tempting place on a landing page to invent — a testimonial here, a customer count there — and the most damaging place to get caught. This prompt builds the trust layer without fabricating any of it: a trust section, clearly-labelled proof placeholders, prompts that tell you exactly what testimonial or metric to collect, risk-reversal copy, an objection-handling FAQ, and a list of every claim that still needs evidence. It works from the real proof and objections you paste; where proof is missing it produces a placeholder describing what to gather, never a fabricated quote, number, award, or guarantee. It produces trust and objection copy, not a full page, an A/B analysis, or invented case studies.
Use only real proof
The trust copy is built from the proof you actually have — nothing is fabricated to fill the section.
Turn gaps into a collection list
Every missing proof point becomes a labelled placeholder telling you what testimonial or metric to gather.
Answer objections honestly
Each real objection gets a straight answer that doesn't overpromise or hide a limitation.
Missing proof becomes a labelled placeholder, never a fabricated quote. Where you write "none yet" for a proof point, the markdown-output-builder returns a PROOF PLACEHOLDERS entry like [TESTIMONIAL: outcome-focused quote from a freelance-designer customer] describing what to gather and where it goes, plus TESTIMONIAL PROMPTS, the exact questions to ask a customer. You collect and paste the real proof yourself.
The CLAIMS NEEDING EVIDENCE section marks every result-or-fact statement as [have it] or [needs evidence], so gaps are visible instead of silently asserted. RISK REVERSAL similarly stays [needs a real guarantee to state] until you supply a real one. The markdown-output-builder produces the markdown; verifying each claim and gathering the proof is your job before you publish.
It writes only the trust layer: the TRUST SECTION, RISK REVERSAL, and OBJECTION FAQ, not the hero or benefits copy. The prompt targets social proof and objection handling specifically, since that is where invention does the most damage. For the hero and benefits blocks, the notFor points to the separate Hero and Benefits copy prompts; this result covers only the sections above.
The contract that stops AI documents from restructuring themselves: a pinned section skeleton, forced tables, and strict consistency rules.
Overview, Installation, Usage, Examples, Configuration — the README skeleton with required, runnable code examples.
Overview, Authentication, Endpoints, Error Handling, Rate Limits — endpoint docs in an identical structure, with parameter tables and runnable examples forced.
Summary, Why It Matters, What Happens Next — the executive summary contract for readers who will never open the source.
Stop getting 'Sure, here is the JSON…' — the output-contract pattern that forces models to return only parseable JSON: schema, example, and a strict rule block.
The JSON won't parse and you can't see why. Deterministic cause-sniffing — trailing commas, single quotes, unclosed brackets — and the repair prompt that fixes it.
Build prompts that produce documents in a fixed structure — headings, sections, and tables.
Write a landing page that converts, not one that just describes — sharpen the value proposition, draft the hero and benefits, answer objections at the CTA, then A/B the variants to pick the stronger.