SEO Engine

The SEO Engine provides a standardized methodology for performing deterministic SEO audits using pre-defined rules and specialized diagnostic scripts. It handles everything from input preparation (fetching live data) to heuristic analysis of spam policies.

Use Case

Perform technical SEO audits on websites or local files to ensure compliance with search engine guidelines and identify potential ranking issues.

Triggering Logic

Use this skill when a user:

Provides a website URL for analysis.
Uploads/references HTML, robots.txt, or sitemap files.
Asks for a "technical SEO check" or "audit".
Specifically mentions SEO-related concerns like "indexability", "crawlability", or "spam signals".

When NOT to use

For general SEO strategy advice (e.g., "how do I rank for 'best laptops'?").
For off-page SEO (backlink analysis, domain authority).
For live performance monitoring (use Lighthouse/CWV tools).

Workflow: SEO Audit Pipeline

Follow these steps sequentially. Do not skip validation gates.

Step 1 — Input Identification & Preparation

Identify Input: URL or local files (@filename.html).
Validation Gate (Input Type): Ensure the input is a valid URL or supported file types (.html, .htm, robots.txt, .xml).
Prepare Live Data (if URL):
- cd scripts/prepare_input/
- Run python fetch_html.py <URL>
- Run python fetch_robots_txt.py <URL>
- Run python fetch_sitemap.py <URL>
Validation Gate (File Presence): Confirm that the fetched files (or provided local files) are readable and non-empty.

Step 2 — Basic Rule Application (Deterministic)

Scan the rules/ directory for applicable rules.
Read the YAML frontmatter of each rule to match inputFields with your prepared data.
Execute Checks: Compare the content of your files against the "Incorrect/Correct" examples in the rule documentation.
Apply Logic: Categorize findings into Pass/Fail.

Step 3 — Specialized Script Execution (Heuristic & Complex)

For complex checks, execute the dedicated Python scripts in the scripts/ directory:

Task	Script Path
Cloaking	`scripts/cloaking_detection/cloaking_detection.py`
Hidden Text	`scripts/hidden_text_detection/hidden_text_detection.py`
Keyword Stuffing	`scripts/keyword_stuffing_detection/keyword_stuffing_detection.py`
Sneaky Redirects	`scripts/sneaky_redirect_detection/sneaky_redirect_detection.py`
Favicon Audit	`scripts/favicon_dimensions/favicon_dimensions.py`
Experience Diversity	`scripts/page_experience_diversity/page_experience_diversity.py`

Instructions for running each script are found in their respective README.md files.

Step 4 — Result Synthesis & Reporting

Normalize Results: Translate pass/fail outputs from rules and JSON results from scripts into a unified report.
Prioritization: Sort results by "Critical", "High", "Medium", and "Low" priorities.
Remediation: For every failure, provide the specific "Actionable Fix" found in the rule documentation or script output.

Example Invocations

Example 1 — Website URL Audit

User: "Analyze https://example.com for SEO issues" Action: Prepare inputs using fetch_*.py -> Apply Rules -> Run specialized scripts -> Generate prioritized report.

Example 2 — Local File Audit

User: "Check @index.html and @robots.txt for SEO compliance" Action: Validate files -> Apply matching rules from rules/ -> Report findings.

Example 3 — Spam Detection

User: "Check if this page is using hidden text or cloaking: @page.html" Action: Run cloaking_detection.py and hidden_text_detection.py -> Report similarity scores and hidden elements.

Troubleshooting

Skill Undertriggering

Cause: Vague request like "Check my site".
Fix: Ask the user to provide a specific URL or upload files. Re-trigger with "I will now run an SEO audit on [URL]".

Preparation Failures

If fetch_html.py fails due to bot detection, advise the user that the site might be blocking headless browsers.
If fetch_robots_txt.py returns 503, immediately flag the ROBOTS_TXT_NOT_503 rule as Critical Fail.

seo-engine