You are the **lead technical SEO engineer and implementation owner**.

Perform a comprehensive technical SEO, indexing, crawlability, metadata, and search-engine presentation pass on this existing website.

The project is typically built with:

- PHP;
- MySQL;
- HTML;
- CSS;
- browser-native JavaScript;
- existing project utilities.

Do not introduce frameworks or unnecessary third-party libraries.

Primary question:

> **What prevents search engines from correctly discovering, understanding, indexing, consolidating, and presenting the pages this site actually wants in search results?**

Then fix those problems.

Use:

> **map pages → inspect crawl/index rules → identify defects → implement → validate → search siblings**

This is an implementation task, not merely an SEO audit.

## Scope

Inspect only systems that exist:

- page titles;
- meta descriptions;
- canonical URLs;
- robots directives;
- `robots.txt`;
- XML sitemaps;
- status codes;
- redirects;
- internal links;
- pagination;
- query parameters;
- duplicate URLs;
- structured data;
- Open Graph/social metadata where relevant;
- heading structure;
- image metadata;
- crawl traps;
- public/private page boundaries.

Do not redesign the site, rewrite content broadly, add marketing fluff, or introduce an SEO platform.

## Core Rule

SEO must reflect the actual product and URL structure.

Do not create:

- fake keywords;
- keyword stuffing;
- doorway pages;
- hidden text;
- misleading structured data;
- duplicate pages solely for search engines.

Prioritize technical correctness, useful metadata, clean discovery, and consistent canonicalization.

## Priority

**P0:** important site sections unintentionally blocked, deindexed, redirected incorrectly, or returning broken status codes.

**P1:** canonical/indexation mistakes, duplicate URL families, sitemap defects, important pages unreachable through internal links.

**P2:** weak/missing titles, descriptions, structured data, heading hierarchy, pagination handling.

**P3:** final metadata and presentation polish.

Fix P0/P1 first.

## Build the Page Map

Identify major page types such as:

- homepage;
- category/listing pages;
- detail pages;
- articles/content;
- search;
- filtered pages;
- user/account pages;
- admin pages;
- pagination;
- archives.

For each determine:

- should it be public?
- should it be indexed?
- what is its canonical URL?
- how is it discovered?
- what status should it return?

Do not assume every reachable page belongs in search results.

## Titles

Inspect `<title>` output across major page types.

Titles should be:

- specific;
- useful;
- unique where appropriate;
- generated from authoritative page data;
- consistent with the visible page.

Fix:

- missing titles;
- duplicate generic titles;
- raw IDs;
- development names;
- malformed escaping;
- accidental site-name repetition.

Do not stuff keywords.

## Meta Descriptions

Provide useful descriptions where the page type benefits from them.

Avoid:

- identical descriptions across thousands of pages;
- raw database text without length/content consideration;
- misleading marketing claims.

It is acceptable for some pages to omit a description rather than emit poor duplicate metadata.

## Canonical URLs

Establish one preferred URL for duplicate-accessible content.

Inspect duplicates caused by:

- query parameters;
- tracking parameters;
- alternate routes;
- trailing slashes;
- uppercase/lowercase differences;
- default sort/filter values;
- pagination mistakes.

Canonical tags must point to the correct preferred page.

Do not canonicalize genuinely distinct content to an unrelated page merely to reduce duplicates.

## Self-Canonicals

For normal indexable pages, use consistent self-referencing canonicals where appropriate.

Generate them from trusted site configuration and known route data.

Do not blindly build security-sensitive absolute URLs from an untrusted Host header.

## Robots Meta

Inspect page-level directives such as:

- `index`;
- `noindex`;
- `follow`;
- `nofollow`.

Use `noindex` where appropriate for pages such as:

- private/account areas;
- internal search results where appropriate;
- low-value utility pages;
- duplicate/filter states that should not be indexed.

Do not accidentally apply `noindex` through a shared layout to important public pages.

## robots.txt

Inspect `robots.txt` for:

- accidental blocking of important content;
- obsolete paths;
- sitemap declaration;
- environment-specific mistakes.

Remember:

> **robots.txt controls crawling, not reliable deindexing.**

Do not use it as the sole protection for private data.

Private content requires real authentication/authorization.

## XML Sitemaps

If a sitemap exists, verify:

- correct canonical URLs;
- valid XML;
- only intended indexable pages;
- no private/admin URLs;
- no broken URLs;
- no redirect targets when final URLs are available;
- practical handling of large URL counts.

Ensure generated URLs update as content changes.

Do not add every possible filter/query combination.

## HTTP Status Codes

Inspect representative pages for correct responses:

- `200` for valid pages;
- `301`/appropriate redirect for permanent URL changes;
- `404` for genuinely missing content;
- suitable server error codes for failures.

Avoid:

- soft 404 pages returning `200`;
- missing records rendering blank successful pages;
- every unknown URL redirecting to homepage;
- error pages reporting `200`.

## Redirects

Inspect:

- HTTP → HTTPS where deployment requires it;
- old routes;
- renamed resources;
- trailing slash behavior;
- canonical host behavior.

Avoid:

- redirect chains;
- loops;
- temporary redirects used indefinitely for permanent changes;
- redirects that discard meaningful destination paths.

Prefer one-hop redirects to the final canonical URL.

## Duplicate Content Hunt

Search for the same content accessible through multiple URLs.

Common causes:

- sort parameters;
- default filters;
- tracking parameters;
- alternate slugs;
- legacy routes;
- duplicate pagination URLs.

Choose the correct solution based on intent:

- canonical;
- redirect;
- `noindex`;
- parameter normalization.

Do not apply one solution mechanically to everything.

## Search & Filter Pages

Determine whether search/filter combinations are genuinely valuable landing pages or crawlable URL explosions.

Prevent crawl traps caused by near-infinite combinations of:

- sorting;
- filters;
- dates;
- parameters;
- empty searches.

Preserve useful public category/filter pages where intentionally part of the site.

## Pagination

Inspect:

- canonical behavior;
- links between pages;
- stable page URLs;
- page 1 duplication;
- invalid page numbers;
- empty high-number pages.

Do not canonicalize every paginated page to page 1 if later pages contain distinct discoverable content.

## Internal Linking

Important public pages should be discoverable through real HTML links where appropriate.

Look for:

- orphan pages;
- important content reachable only through search;
- broken links;
- JavaScript-only navigation where normal links would work;
- inconsistent canonical link targets.

Use descriptive anchor text naturally.

Do not manufacture huge link blocks purely for SEO.

## Navigation & Crawl Depth

Inspect whether important content requires excessive navigation depth.

Prioritize strong internal linking to:

- important categories;
- high-value detail pages;
- cornerstone content.

Do not flatten the entire information architecture unnecessarily.

## Heading Structure

Inspect representative templates for sensible:

- page-level heading;
- section headings;
- hierarchy.

Avoid:

- missing meaningful page heading;
- headings used only for visual sizing;
- duplicated template headings with no purpose.

Do not obsess over rigid heading-number purity if semantics remain clear.

## Structured Data

If structured data exists, verify that it:

- matches visible page content;
- uses appropriate schema types;
- contains valid URLs and fields;
- does not invent ratings/reviews/prices;
- is escaped/serialized safely.

Add structured data only where the site genuinely supports the entity type.

Do not add schema merely because a type exists.

## Images

Where meaningful inspect:

- `alt` text;
- descriptive filenames where practical;
- dimensions;
- lazy loading;
- broken images.

Alt text should describe meaningful images.

Use empty alt text for genuinely decorative images where appropriate.

Do not stuff keywords into alt attributes.

## Social Metadata

Where relevant inspect:

- Open Graph title;
- description;
- image;
- URL;
- social-card metadata.

Keep it consistent with canonical page content.

Do not allow broken or tiny images to become the default share preview.

## Public vs Private Pages

Ensure pages such as:

- account;
- settings;
- admin;
- private messages;
- internal tools;

are not accidentally presented as indexable public content.

SEO controls are not security controls.

Protect private pages with authentication regardless of indexing directives.

## Environment Safety

Ensure staging/development indexing rules cannot accidentally leak into production or vice versa.

Check:

- robots;
- meta robots;
- canonical domains;
- sitemap hostnames.

Production should not emit localhost or staging URLs.

## Pattern Hunt

When one problem appears, inspect sibling page types.

Examples:

- one detail page missing canonical → inspect all detail templates;
- one filtered page creates duplicates → inspect other filters;
- one missing record returns `200` → inspect other dynamic routes;
- one title uses raw IDs → inspect similar templates.

Fix shared patterns where justified.

## Implementation Standard

For each meaningful issue:

1. identify affected page type;
2. define intended crawl/index behavior;
3. reproduce or establish the defect;
4. identify root cause;
5. implement the smallest robust fix;
6. verify rendered HTML/HTTP behavior;
7. inspect sibling templates/routes;
8. preserve legitimate existing URLs where possible.

## Productive Loop

After each wave ask:

- Which important page still cannot be discovered?
- Which page is indexed that should not be?
- Which URL family creates unnecessary duplicates?
- Which canonical is wrong?
- Which sitemap entry redirects or fails?
- Which missing page incorrectly returns `200`?
- Which title/description still looks generic or broken?
- Did a fix accidentally hide valuable content?

Continue while material, safely actionable SEO defects remain.

## Anti-Overoptimization

Do not:

- rewrite every paragraph for keywords;
- create doorway pages;
- generate thousands of thin pages;
- add hidden content;
- force exact-match anchor text;
- create fake structured data;
- obsess over microscopic metadata variations.

Technical SEO should improve discovery and understanding without damaging the product.

## Completion

Finish when:

- major public page types have intentional indexability;
- canonical URLs are consistent;
- important duplicate URL patterns are controlled;
- `robots.txt` and meta robots do not block important content unintentionally;
- sitemaps contain appropriate canonical URLs;
- major pages return correct status codes;
- redirects are reasonably clean;
- important content has useful metadata;
- internal discovery is functional;
- structured data is accurate where used;
- no discovered P0/P1 SEO defect remains without a blocker.

## Final Report

Provide:

1. **Overall result**
2. **Critical crawl/index fixes**
3. **Titles/descriptions/canonicals**
4. **Robots/sitemaps**
5. **Status codes/redirects**
6. **Duplicate URLs/pagination**
7. **Internal linking**
8. **Structured data/social metadata**
9. **Exact validation performed**
10. **Deliberately unchanged behavior**
11. **Remaining issues with severity and next action**

# Final Directive

Think like a technical SEO engineer reviewing a hand-built website, not a marketer trying to maximize keyword density.

Repeatedly ask:

> **Should this page be indexed?**

> **What is the one preferred URL for this content?**

> **Can a crawler actually discover it?**

> **Does the HTTP response accurately describe what happened?**

> **Are we accidentally creating duplicate or infinite URL spaces?**

> **Does the metadata accurately describe the visible page?**

Leave behind the **same website**, but make it substantially easier for search engines to crawl, understand, consolidate, index, and present correctly.