calevans/answer-engine-optimization

StaticForge Feature: AnswerEngineOptimization

Maintainers

👁 calevans

Package info

github.com/calevans/staticforge-answer-engine-optimization

pkg:composer/calevans/answer-engine-optimization

Statistics

Installs: 11

Dependents: 1

Suggesters: 0

Stars: 0

Open Issues: 0

2.2.3 2026-06-15 23:43 UTC

Requires

Requires (Dev)

Suggests

None

Provides

None

Conflicts

None

Replaces

None

MIT 79f069ba552962a9a9d430191d3d7b5062b8401f

This package is auto-updated.

Last update: 2026-06-15 23:45:17 UTC

README

A StaticForge feature package that optimises your site for AI-powered answer engines (LLMs, RAG pipelines, AI search). It handles structured data injection, llms.txt generation, AI-bot robots.txt rules, and data-driven FAQ schema — all automatically as part of the StaticForge build pipeline.

Installation

composer require calevans/AnswerEngineOptimization

Features

Feature Description
AI-bot robots.txt Adds Allow: / rules for OAI-SearchBot, ChatGPT-User, Google-Extended, Anthropic-ai, and Claude-Web
Article JSON-LD Injects Article structured data into every page <head>
BreadcrumbList JSON-LD Injects BreadcrumbList structured data on all non-home pages
FAQ JSON-LD Injects FAQPage structured data from frontmatter, [faq] shortcodes, or a shared JSON data file
llms.txt Generates a site-wide llms.txt index for AI crawlers, with a titled header and per-page descriptions
llms link tag Injects <link rel="llms" href="/llms.txt"> into every page <head>
.md copy Publishes a stripped Markdown copy of each page alongside its HTML output

Configuration

siteconfig.yaml

All keys are optional. The package works with sensible defaults when they are absent.

site:
 name: "My Site" # Becomes the `# Title` in llms.txt and the Article schema publisher name.
 # Defaults to "LLMs Documentation" in llms.txt when not set.
 tagline: "My tagline" # Becomes the `> description` blockquote in llms.txt.
 # Falls back to `site.description` if `tagline` is absent.
 description: "..." # Fallback description for llms.txt when `site.tagline` is not set.

social:
 default_image: "https://example.com/logo.png" # Added as `publisher.logo` in the Article JSON-LD schema.

answer_engine_optimization:
 faq_data_file: content/assets/faq.json # Path (relative to project root) to the shared FAQ JSON data file.
 # Defaults to `content/assets/faq.json` when not set.

Per-page frontmatter

---
title: "Page Title"
description: "Shown as the description in the llms.txt entry for this page."
no_llms: true # Set to true to exclude this page from llms.txt and suppress the <link rel="llms"> tag.
aeo:
 key_takeaways: "A custom summary written by the author, used verbatim in llms.txt."
 faqs:
 - question: "What is this page about?"
 answer: "It is about answer engine optimisation."
---

FAQ data file

When answer_engine_optimization.faq_data_file is configured (or the default content/assets/faq.json exists), the package loads a shared bank of FAQs that can be referenced from any page via data-faq HTML attributes.

content/assets/faq.json

[
 {"id": "what-is-aeo", "question": "What is answer engine optimisation?", "answer": "It is the practice of structuring content so AI systems can accurately surface it in answers."},
 {"id": "pricing", "question": "How much does it cost?", "answer": "It is free and open source."}
]

Reference an entry in your Markdown by adding a data-faq attribute to any HTML element:

<div data-faq="what-is-aeo"></div>
<div data-faq="pricing"></div>

The matching entries are automatically included in the FAQPage JSON-LD schema for that page.

Priority order for FAQ sources

FAQs from all three sources are merged in this order:

  1. aeo.faqs in page frontmatter
  2. [faq question="…" answer="…"] shortcodes in page content
  3. data-faq="id" references resolved from the shared JSON data file

Priority order for llms.txt summary

The per-page summary line in llms.txt is resolved in this order:

  1. aeo.key_takeaways in page frontmatter (author-written, highest priority)
  2. description in page frontmatter
  3. Empty string (the colon is still emitted; the AI crawler can infer context from the .md copy)