agentleFS
frontmost-aifrontmost-ai/aeo-skills/skills/answer-pages/SKILL.md

answer-pages

Skill1 starsChanged 3 days agoMITView on GitHub

Plan a page on the business's own site from real AI answers: a page for one buyer question, a fair comparison page, a better version of a page that AI apps already cite, or page ideas from the web searches that ChatGPT ran. Use before you write content for AI search. Needs a Frontmost AI project.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If a tool returns setup_required, give the person the url from the result.

    line 14
  2. Very specific:

    If a tool returns plan_required, say that this data needs a paid Frontmost AI plan, and stop.

    line 15
  3. Very specific:

    If the Frontmost AI tools are not available, use the page-drafting skill.

    line 16
  4. Distinctive:

    Never invent the data, and never fill a gap with a guess.

    line 17
  5. Very specific:

    Call the tool for the job: get_questions for a page for one question, get_rivals for a comparison page, get_sources for pages that AI apps …

    line 21
  6. Distinctive:

    Write down three things: the points that the answers repeat, the facts that they use to compare businesses (price, features, limits, who …

    line 23
  7. Distinctive:

    Facts: use only facts that the person confirms.

    line 30

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Answer pages

Use this guidance to plan a page on the business's own website from the answers that AI apps gave. It covers four jobs: a page for one question, a comparison page, improvements to pages that AI apps already cite, and page ideas from ChatGPT's searches.

Before you start

This skill reads a Frontmost AI project through the Frontmost AI tools that this plugin connects.

  • If a tool returns setup_required, give the person the url from the result. Stop until they say that set-up is done.
  • If a tool returns plan_required, say that this data needs a paid Frontmost AI plan, and stop.
  • If the Frontmost AI tools are not available, use the page-drafting skill. It needs no account. To plan pages from real AI answers, the person connects Frontmost AI: https://frontmost.ai/connect
  • Never invent the data, and never fill a gap with a guess.

Read the answers first

  1. Call the tool for the job: get_questions for a page for one question, get_rivals for a comparison page, get_sources for pages that AI apps already cite, and get_queries for page ideas from ChatGPT's searches.
  2. Open the real answers. Use question_id, rival_id, or the row's answers_filter in get_answers. Do not build a filter.
  3. Write down three things: the points that the answers repeat, the facts that they use to compare businesses (price, features, limits, who the product fits), and the format of the pages that they cite (comparison, list, guide, review).

Plan the page

  • Title: the buyer's question in plain words.
  • First paragraph: a direct answer in 2 or 3 sentences that a reader can quote without the rest of the page.
  • Body: one section for each point that the answers repeat. Put the specific fact first in each section.
  • Facts: use only facts that the person confirms. When you need a price, a limit, or a feature and you do not have it, ask. Never invent one.

Job rules

  • One question: unless the person names one, pick the question with the most answers that name a rival and not the business.
  • Comparison: take each point from a real answer and show the quote. Include the points where the rival is stronger. A page that hides them is not fair, and readers will see it.
  • Cited pages: for each of the business's own pages that AI apps cite, say what the answers took from it and what the answers needed and did not find there.
  • Searches: group ChatGPT's searches by what the buyer wants. Give one page idea for each group. Ask which groups the site already covers before you call a group a gap. When a page on the site already covers the group, add the subject to that page.

Never suggest

  • llms.txt files or "AI schema". Google states that its AI features need no new files and no special markup.
  • Keyword stuffing, hidden text, or copied text from a rival.
  • Any promise that AI apps will cite or mention the page. No method is proven to earn a citation. A page is a bet that the data supports.

Output

An outline: the title, the first paragraph as a draft, the section headings with one line on what each section must state, and the list of facts that you still need from the person.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.