What it does
Reads a page or product and the questions customers actually ask, then assembles an FAQ block with schema-ready JSON-LD markup so it can also earn a rich result in search, without a dedicated FAQ or schema plugin.
Install
This skill isn't packaged in an installable registry. It's a plain SKILL.md file you can copy directly. Copy it from the SKILL.md tab and save it at the path for your assistant:
Use it in your AI assistant
Claude Code
.claude/skills/internal-faq-builder/ (or ~/.claude/skills/internal-faq-builder/ for all projects)
GitHub Copilot (VS Code)
.github/skills/internal-faq-builder/
OpenAI Codex
.codex/skills/internal-faq-builder/
Cursor
.cursor/skills/internal-faq-builder/ (or ~/.cursor/skills/internal-faq-builder/ globally)
Compatibility
Any WordPress site using the block editor. Writing valid FAQPage JSON-LD needs no plugin, just a place to output a script tag (theme template or a code/HTML block).
The complete SKILL.md for this skill:
---
name: internal-faq-builder
description: Use when a user wants an FAQ section added to a page, wants existing support questions turned into an FAQ, or wants FAQ schema markup for search rich results.
---
# FAQ Builder
## When to use
A page or product needs an FAQ section, built either from the page's own content or from a supplied list of real customer questions (support tickets, comments, sales calls).
## Procedure
1. Gather the questions: either extract implicit ones from the page content (things a reader would still wonder after reading) or use the list the user supplies.
2. Write each answer directly from the page's actual content or supplied facts, 2-4 sentences, no filler ("Great question!").
3. Assemble a Q&A block using WordPress core blocks (heading + paragraph pairs, or a details/summary pattern) so it renders without a dedicated FAQ plugin.
4. Generate matching FAQPage JSON-LD (schema.org) with the same question/answer pairs, output as a single script tag the user can place in the page.
5. Cap at 6-8 questions per page; more than that reads as padding and dilutes the schema's relevance.
## Guardrails
- Every answer must be traceable to real page content or a real supplied fact, never invent a plausible-sounding answer to a question that wasn't actually addressed anywhere.
- Keep the JSON-LD and the visible on-page text in sync; mismatched schema is a common cause of lost rich results.