agentleFS
oegeyilmaz9oegeyilmaz9/seo-aeo-geo-ultimate/skills/seo-aeo/SKILL.md

seo-aeo

Skill6 starsChanged 3 months agoApache-2.0View on GitHub

Audit supplied pages, documents, or content sets for direct-answer completeness, question and intent coverage, clarity, extractability, claim support, and visible-content consistency against a pinned AI-search Research Pack. Use for AEO, answer-engine optimization, answer readiness, answer extraction, featured-answer preparation, or requests asking whether content clearly and supportably answers researched questions. Do not use for generative-engine mention/citation strategy, live research, content rewriting, schema generation, technical SEO implementation, or visibility monitoring.

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    Read markdown-delivery-contract.md for every task.

    line 10
  2. Distinctive:

    Read audit-protocol.md before scoring findings or recommending a handoff.

    line 22
  3. Distinctive:

    Verify the Research Pack identity, schema version, hash, declared locales, queries, evidence, ground truth, and gaps.

    line 26
  4. Distinctive:

    Verify every target capture, capture time, target type, and locale.

    line 27
  5. Very specific:

    Create an optimization-brief.json with optimization_domain: "aeo" and producer_skill: "seo-aeo".

    line 39
  6. Very specific:

    Do not perform GEO mention/citation strategy.

    line 55

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “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

SEO AEO

Markdown delivery contract

Read markdown-delivery-contract.md for every task. Write a substantive .md document for every material work step and register its direct destination in the work index.md. For public-site work, apply the owner-required Markdown discovery profile across the full eligible inventory; preserve native standards and immutable evidence formats.

Lane outputs: aeo-findings.md; answer-map.md (combine small related steps with stable section links). Audit substantive Markdown answers alongside their pinned HTML/source captures: self-contained meaning, question-to-section access, completeness, dates, limitations and HTML parity. Keep formal Research Pack requirements for engine-scoped claims. Route publisher navigation/parity defects to technical/content owners without inventing engine evidence.

Produce an evidence-linked AEO Optimization Brief from immutable inputs. Audit answer readiness; do not implement the remedies.

Required inputs

  • Require a contract-valid research-pack.json from ai-search-research.
  • Require a dated, bundle-relative capture plus a hash-pinned target-metadata file containing target ID, type, locale, supplied source URL, capture reference, capture SHA-256, and capture time. Never trust Brief fields without matching metadata, invent a canonical URL, or audit an uncaptured, unpinned, or silently refreshed page.
  • Record inaccessible, incomplete, stale, or locale-mismatched inputs as limitations. Do not infer missing content.
  • Read audit-protocol.md before scoring findings or recommending a handoff.

Procedure

  1. Verify the Research Pack identity, schema version, hash, declared locales, queries, evidence, ground truth, and gaps.
  2. Verify every target capture, capture time, target type, and locale. Stop with an explicit gap if a required capture is missing.
  3. Audit only these AEO dimensions:
    • direct-answer completeness;
    • question, intent, audience, and journey-stage coverage;
    • answer clarity and self-contained meaning;
    • claim-to-evidence support and freshness;
    • entity and terminology consistency within the answer;
    • visible-content consistency with any supplied structured-data representation.
  4. Verify every Brief target field against the hash-pinned metadata file. Record target observations as intra-brief direct_observation evidence pinned to the target capture and metadata-supplied source URL. Do not invent a URL or republish Research Pack source records as audit evidence. A target finding must cite that observation; research-dependent conclusions must also cite scope-matched Research Pack evidence.
  5. Bind each finding to resolved Research Pack query IDs. Keep the finding locale equal to its target locale and keep engine/surface scope within those queries' applicability.
  6. Classify each observation as confirmed, vendor-recommended, experimental, or speculative. A finding or confidence level cannot exceed its weakest necessary premise. Speculative evidence never supports an implementation recommendation. Experimental-only support belongs in experiments; route unsupported premises to limitations, declined claims, or research backlog.
  7. Reject unsupported formulas, word-count thresholds, guaranteed placements, universal crawler claims, and universal engine-mandatory llms.txt claims into declined_claims.
  8. Create an optimization-brief.json with optimization_domain: "aeo" and producer_skill: "seo-aeo". Pin every external evidence reference to the immutable Research Pack hash. Use SELF only for an intra-brief audit_evidence reference. Keep stable IDs unique and never copy Research Pack evidence into audit_evidence.
  9. Give every recommendation its own evidence references, desired outcome, confidence, verification method, and exactly one candidate implementation owner. Its evidence must overlap every linked finding and remain scope-compatible. A Brief does not approve a change.
  10. Set created_at after all target captures and audit observations. Validate Research Pack uniqueness, chronology, capture hashes, query applicability, evidence resolution and scope, and the Optimization Brief contract before delivery.

Ownership boundaries

Route implementation planning first; never perform or approve implementation inside this skill:

  • seo-action-plan: evidence-linked ownership, approval gate, verification, and rollback for a validated Brief.
  • seo-content: substantive wording, editorial quality, or answer rewrite.
  • seo-schema: structured-data choice, generation, syntax, eligibility, or validation.
  • seo-technical: crawlability, rendering, indexing, directives, performance, or transport controls.
  • seo-hreflang: locale and regional mapping.
  • optimise-seo: broader integrated page optimization outside the AEO delta.

Do not perform GEO mention/citation strategy; route that to seo-geo. Do not measure live visibility or drift; route that to ai-visibility-monitor. Do not create missing research; route that to ai-search-research.

Severity and review

  • critical: the audit or handoff would be materially false, unsafe, or unusable without correction.
  • important: a required answer, evidence link, locale, or owner handoff is missing or misleading.
  • minor: non-blocking clarity or maintainability improvement; record it in backlog.

Close all critical and important audit findings. Backlog minor findings.

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.