nexscope-ai/ecommerce-seo-geo-skills / ecommerce-geo-auditor
ecommerce-geo-auditor
VerifiedDiagnose an ecommerce page or listing for GEO, AI-search, AI-citation, and AI-shopping readiness using inspected page evidence, supplied content, or crawl data. Use when the user asks for a GEO audit, AI visibility readiness check, AI search audit, citation-readiness review, or why a product is hard for AI systems to understand. Diagnose only; do not claim actual AI mentions or rewrite the full page.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.4 KB |
| agents/openai.yaml | config | 305 B |
| manifest.json | config | 500 B |
| references/audit-checklist.md | markdown | 1.4 KB |
| references/geo-pillars.md | markdown | 3.9 KB |
| references/schema-templates.md | markdown | 1.0 KB |
references/geo-pillars.md
View on GitHub# Ecommerce GEO readiness framework
Use this framework as a qualitative diagnostic aid, not as a claim about proprietary AI ranking factors. Readiness and measured AI visibility are different outcomes.
## Contents
1. Access and search foundation
2. Product facts and identity
3. Structured data and consistency
4. Buyer-intent coverage
5. Trust and proof
6. Answerability and content structure
7. Media and page experience
8. External visibility evidence
9. Optional heuristic scoring
## 1. Access and search foundation
Inspect only with live or technical evidence:
- Page is accessible, indexable, canonicalized appropriately, and linked internally.
- Important text is present in rendered HTML.
- Search controls do not unintentionally block the page.
- Page experience and mobile rendering do not prevent use.
Do not infer these checks from copy alone.
## 2. Product facts and identity
Look for:
- Clear product, brand, category, audience, and use case
- Visible specifications, variants, price, availability, and policies
- Consistent facts across the page and structured data
- Important information available as text, not only images
There is no universal requirement that a brand appear within an exact word count.
## 3. Structured data and consistency
Look for accurate Product and Offer-family markup where the platform permits it. Confirm that values match visible content.
Structured data can help systems understand a page and enable eligible search features. It does not guarantee rankings, rich results, AI citations, or recommendations. Google states that its AI search features require no special AI schema beyond normal search eligibility and accurate structured data.
## 4. Buyer-intent coverage
Assess whether the page answers relevant questions about:
- Fit and audience
- Use cases and limitations
- Specifications and compatibility
- Price or value context
- Alternatives and differentiators
- Shipping, returns, warranty, and support
Use genuine demand, support, review, or query evidence when available.
## 5. Trust and proof
Check for visible, verifiable evidence such as:
- Genuine reviews and ratings
- Clear seller and brand identity
- Accurate shipping, return, warranty, and contact information
- Certifications or tests with traceable sources
- Consistent third-party evidence when supplied
Never create or embellish trust signals.
## 6. Answerability and content structure
Useful patterns include:
- Direct explanations of what the product is and who it serves
- Specific facts in clear sentences and tables
- Descriptive headings
- Concise answers to genuine buyer questions
- Honest limitations and comparison criteria
FAQ formatting can improve usability, but it is not a universal citation mechanism. Google FAQ rich results are generally limited to eligible authoritative government and health sites.
## 7. Media and page experience
Review:
- Relevant crawlable images and video
- Descriptive alt text that serves accessibility
- Responsive delivery and observed performance evidence
- Text alternatives for important information shown in media
Do not use image filenames, WebP, or an arbitrary file-size threshold as a standalone AI-visibility score.
## 8. External visibility evidence
Actual visibility requires dated, query-level evidence such as:
- Whether the product was mentioned
- Recommended position or role
- Citations or source URLs
- Competitors shown for the same query
- Model, market, language, and run time
Do not infer actual mentions from page readiness alone.
## 9. Optional heuristic scoring
If the user explicitly wants a score and enough evidence is available, score each area:
- `0` — weak or blocked with confirmed evidence
- `1` — partial or mixed
- `2` — strong with confirmed evidence
- `Unknown` — not inspected
Report the number of scored and unknown areas. Do not convert unknowns to zero, forecast a future score, or claim a score predicts citations, traffic, or sales.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |