AEO Foundation

This skill builds the foundation file that every other AEO skill reads first: aeo-foundation.md. Answer Engine Optimization lives or dies on context — to influence what ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews say about a brand, an agent has to know the brand as an entity, the prompts buyers actually ask, the competitors who get recommended instead, and where the brand stands today. Captured once, this file makes audits sharper, content on-target, and authority work focused. It works for any business, B2B or B2C, product or service.

The output is a single Markdown file — aeo-foundation.md — kept at the root of the user's AEO workspace. Other skills look for it by name; if it is missing, they should point the user here.

When to use this skill

• Starting AEO work on a new brand, product, service, or client.
• Another AEO skill needs the prompt set, entity facts, competitors, or baseline and none exists.
• The context is stale — a new competitor emerged, positioning shifted, the brand was misrepresented by an engine, or the prompt set has drifted.

How to build it

Work like an interviewer, not a form. Pull everything you can from what the user already has — their website, About page, existing schema, review profiles, prior answers — then ask only for what is genuinely missing, a few questions at a time.

Completeness is optional. The user does not need to answer everything, or anything. Capture what you can, tag gaps with (assumption) so later work knows what to verify, and produce a useful file regardless. A thin foundation today beats a perfect one never; the file is designed to grow as the engagement learns.

A concrete first move that grounds everything else: run a quick baseline. Ask each major engine a handful of the brand's likely prompts and record what they say, who they cite, and whether the brand appears at all. That snapshot anchors the whole strategy.

Integrating an existing context

Many brands already have a context document — a positioning one-pager, a brand/voice guide, an ICP doc, or a general marketing context file. Reuse it; do not start from scratch and do not blindly copy it. Best practice:

• Reference, don't duplicate. Keep one source of truth per fact. The existing doc stays upstream for the general "who/what/why"; aeo-foundation.md is a thin AEO layer on top of it. Two patterns:
  • Link (for living docs the user maintains): point to the existing file for the shared sections and only fill the AEO-specific ones here. No drift.
  • Import once (for static/one-off docs): map its fields into the shared sections, note provenance and date, and maintain forward. Add a "re-sync when the source changes" note.
• Know what carries over vs. what is AEO-only. Reuse offering, audience/personas, positioning, differentiators, voice, proof, and the seed competitor list. The AEO-specific layers a normal context doc lacks must still be built: the target prompts (conversational questions, via prompt-research), the AI-visibility baseline, sharper entity facts, and the technical context.
• Sharpen, don't transplant. AEO needs more specificity than a brand deck: exact, web-consistent entity facts and the literal questions people ask AI. Reconcile any conflicting facts toward what is true on the public web. Treat the user's competitor list as a seed, then verify who engines actually recommend in the baseline.

Gather areas

Cover as many as the user can provide. Each is optional.

1. Brand entity — official name (and disambiguation if the name is generic), category, one-line description, founding date, headquarters, key people, and the canonical profile URLs (site, LinkedIn, Crunchbase, Wikipedia/Wikidata if any).
2. Offering — what it is and does, the category customers file it under, core use cases and the jobs it gets done.
3. Audience & personas — who asks AI about this category, their context, constraints, and the language they use; distinct segments ranked by priority.
4. Target prompts — the real conversational questions buyers ask AI across awareness, consideration, and decision intent. This becomes the tracked prompt set the other skills optimize and measure against. Aim for a focused set of high-value prompts over a sprawling one.
5. Competitor / alternative set — who else gets recommended for those prompts, including the status quo, and the comparison points that matter.
6. Positioning & differentiators — the associations you want engines to make with the brand, and any unique mechanism or credible "only" claim.
7. Authority assets — original data or research, named experts and authors, third-party coverage and press, analyst mentions, review-platform presence, and notable corroboration that already exists.
8. Current AI-visibility baseline — per engine, what is said about the brand today, whether it is cited or merely mentioned, the sentiment, who out-cites it, and any factual errors to correct.
9. Voice & boundaries — claim constraints, regulated or sensitive terms, compliance needs (especially for health, finance, legal / YMYL topics).
10. Technical context — site platform and rendering (server-side vs client-side), CMS, whether AI crawlers are currently allowed, and the state of structured data.

Output template

Produce the file using the structure in templates/aeo-foundation-template.md. Omit a section only if it truly does not apply; prefer a short (assumption)-tagged draft over a blank. Keep it tight enough to read in a few minutes — it is a working reference, not a brand bible.

Maintenance

Treat the file as living. When other AEO work surfaces something — a prompt that converts, a competitor that keeps winning citations, a misrepresentation an engine keeps repeating, a new proof asset — update the foundation file and bump the date. The whole library compounds off this file staying current; re-run the baseline periodically to track movement.

Related Skills

• prompt-research — expands and prioritizes the target prompt set captured here.
• ai-visibility-audit — turns the baseline into a scored, prioritized audit.
• competitor-ai-analysis — deepens the competitor/alternative picture.
• entity-optimization — acts on the brand-entity facts recorded here.
• All content, authority, technical, platform, and measurement skills read this file for audience, prompts, positioning, and proof.