answer-block-generator-aivis

name: answer-block-generator-aivis description: > Generates citation-optimized answer blocks, FAQ sections, and direct-answer content units specifically structured to score well on the AiVIS Cite Ledger AI Readability dimension (12% of composite score). Each output is a self-contained, verifiable, extractable content unit that AI answer engines can quote without surrounding context. Produces entity definition paragraphs, speakable-ready blocks, FAQ Q&A pairs, and methodology step content. Use this skill whenever the user needs to write or rewrite content that AI engines will quote, generate FAQ sections, create answer-ready content for a product or service page, or improve their AI Readability score. Trigger on "write a FAQ", "make this answer-ready", "content AI engines will cite", "speakable content", "direct answer blocks", "improve AI readability score", or requests to write content for specific pages like methodology, about, or product feature sections.

Answer Block Generator, AiVIS Cite Ledger AI Readability Framework

Generates content units that maximize the AiVIS Cite Ledger AI Readability dimension score (target: 80+/100 per block).

What makes a block AI-citable

An answer block is citable when it satisfies all four extraction requirements:

1. Self-contained, reads correctly in isolation with no pronouns needing context
2. Specific, contains at least one verifiable fact, number, or named element
3. Concise, answerable within 1–3 sentences; longer answers score lower
4. Entity-anchored, the brand, product, or topic name appears in the answer

Block types and templates

Type 1: Entity definition block (canonical, every homepage)

Purpose: primary extraction target for "What is [brand]?" queries. Speakable schema should point here.

[Brand] is a [specific category] that [verb + what it does] for [audience].
[Named differentiator: what makes it distinct from the default solution in the category].
[One concrete output: what the user receives, with a specific fact or metric].

Example (AiVIS Cite Ledger):

AiVIS Cite Ledger is an AI visibility audit platform that scores how web content is
interpreted, trusted, and cited by AI answer engines including ChatGPT,
Perplexity, Gemini, and Google AI Overviews. Unlike traditional SEO tools,
AiVIS Cite Ledger uses the CITE LEDGER framework to measure citation readiness across
ten scoring families (avs-v3) rather than keyword rankings. Each audit returns a
0–100 composite score with BRAG evidence-linked findings tied to observable
page structure.

Type 2: FAQ Q&A block (highest citability format)

Rules:

• Question must be the exact phrase a user would search
• Answer must be complete in 1–3 sentences
• No "it depends" or "it varies" without an immediate specific follow-up
• No reference to other parts of the page

Template:

### [Question, natural language, includes primary term]
[Brand/product/concept] [direct answer, verb + object + context].
[Optional: one supporting fact or example].

Type 3: Methodology step block

Named steps are extracted and cited more frequently than unnamed steps.

Template:

**[Step N]: [Named stage],** [What the system does at this stage, specifically].
Input: [what goes in]. Output: [what comes out].

Type 4: Scoring or grading block

Tables and tier systems are among the most-cited formats in AI answers. Always anchor thresholds to named, observable criteria.

Template:

| Grade | Score range | Meaning |
|---|---|---|
| [Grade A] | [X–Y] | [Specific outcome in observable terms] |
| [Grade B] | [X–Y] | [Specific outcome] |

Type 5: Comparison block

"X vs Y" content is highly cited. Structure as direct parallel statements, not narrative paragraphs.

Template:

**[Option A]:** [Specific attribute, number or named feature].
**[Option B]:** [Same attribute for comparison, number or named feature].
[One sentence conclusion that is citable as a standalone claim].

Quality gates

Run every block through these before outputting:

• Can this be quoted without the surrounding page? (self-contained test)
• Does it contain a specific fact, number, or named element?
• Is the brand or topic name explicitly stated (no "it" or "they")?
• Is the answer under 100 words? (if not, split into two blocks)
• Would this answer a specific search query directly?

AI Readability scoring signals targeted

Speakable integration

After generating blocks, identify which ones should be marked speakable:

• Entity definition block: always speakable
• Top 3 FAQ blocks: speakable candidates
• Methodology step 1 (what the system does): speakable

Output the cssSelector targets alongside content:

Content: [block text]
Speakable target: #[suggested-id] or .[suggested-class]
Schema: { "@type": "SpeakableSpecification", "cssSelector": ["#[id]"] }

Output format

For each request:

1. Generate the content block(s), ready to publish
2. Label the block type (Entity / FAQ / Methodology / Scoring / Comparison)
3. Note which AiVIS Cite Ledger AI Readability signals each block satisfies
4. Flag any speakable candidates
5. Estimated AI Readability contribution if this replaces existing content