answer-readiness
Skill0 starsChanged 49 days agoMITView on GitHub
Restructure a page so an AI engine can lift an answer out of it. Use when writing or editing content for AI search, when a page ranks but is never cited, or when someone asks how to structure content for ChatGPT, Perplexity or AI Overviews.
In this repository4 other files it gives its agents.
- ai-visibility-checkSkill
- answer-gap-contentSkill
- geo-auditSkill
- llms-txtSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Write the opening so it survives being cut out and pasted somewhere else.
- Distinctive: line 24
Use the question as the heading where that reads naturally, not a keyword fragment.
- Distinctive: line 27
Avoid headings that only make sense in sequence ("The second problem").
≈ “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.
Files
Answer readiness
A page gets cited when a model can find a self-contained answer inside it and quote that answer without reading the whole page. Most pages that fail this are not badly written; they are badly shaped.
The shape that gets quoted
Answer first, in the first hundred words
State the answer outright, before any context, story or preamble. If the question is "how much does X cost", the number goes in the opening sentence.
Engines quote the first direct answer they find. A page that spends three paragraphs establishing why the question matters hands the citation to the page that answered in its first line.
Write the opening so it survives being cut out and pasted somewhere else. That is literally what happens to it.
One heading per question people actually ask
Headings are the retrieval unit. A model matching a question against your page matches it against headings first.
- Use the question as the heading where that reads naturally, not a keyword fragment.
##for sections,###for sub-questions. Do not skip levels.- One
<h1>. - Avoid headings that only make sense in sequence ("The second problem"). Each one is read alone.
Put something liftable in the middle
The formats engines pull whole:
- A comparison table. One row per option, one column per attribute, units in the header. This is the single most quoted structure in buying questions, because the model can lift it without paraphrasing.
- An FAQ. Real questions with short direct answers, each question as a heading. Cover the ones the body did not.
- A definition or a numbered procedure. Self-contained, no "as mentioned above".
A page of unbroken prose can be entirely correct and still never be quoted, because there is nothing in it shaped like an answer.
Say who says it
Attribution and specificity are what make a passage safe to quote: a named author, a date, a stated method, a real number with its source. "Studies show" is unquotable. "In our 2026 test of 69 questions across 14 niches, the median was 96%" is quotable, because a model can attribute it.
Checking a page
Score the page against these and report what is missing with the fix beside it:
- Is the answer in the first hundred words, or is there preamble first?
- Does any heading match a question a buyer would type?
- Is there a table, an FAQ or a procedure, or is it all prose?
- Can any single section be read alone and still make sense?
- Is there a specific, attributable number or claim anywhere?
- Does the title tag and meta description state what the page answers, cut at a sentence rather than mid-clause?
Two things that do not work
- Keyword density. It is not a retrieval signal for these engines and chasing it costs you the readability that is.
- Writing for a model rather than a person. The pages that get quoted read well to a human. Content that is obviously shaped for a machine tends to be thin, and thin content does not survive the comparison a model does between its sources.
With the Peak Answer MCP connected
audit_site returns an answer-readiness score per page alongside the technical checks. generate with content_brief produces a brief already in this shape, and with faq_schema produces valid FAQPage JSON-LD from a page's real questions.
Free, no signup: https://peakanswer.com/tools/answer-readiness-score
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
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.

