EEAT Content Quality Audit

This skill is developed based on the CORE-EEAT Content Benchmark, providing 80 standardized content quality audit criteria.

Skill Overview

This skill evaluates content quality through 80 standardized criteria across 8 core dimensions. It generates comprehensive audit reports including item-level scores, dimension scores, system scores (GEO/SEO), content-type weighted total scores, veto item detection, and priority action plans.

Applicable Scenarios

Use this skill when users request the following:

• - Content quality check or audit
• EEAT scoring or E-E-A-T audit
• Content quality assessment
• CORE-EEAT audit
• GEO quality scoring
• Content improvement recommendations
• AI citation potential assessment
• Content optimization plan
• "How good is my content"
• "Can my content be cited by AI"

Core Capabilities

This skill can:

1. 1. Complete 80-Item Audit: Score each CORE-EEAT item as Pass/Partial/Fail
2. Dimension Scoring: Calculate scores for all 8 dimensions (0-100 points each)
3. System Scoring: Calculate GEO score (CORE) and SEO score (EEAT)
4. Weighted Total Score: Calculate final score based on content-type specific weights
5. Veto Item Detection: Flag critical credibility violations (T04, C01, R10)
6. Priority Ranking: Identify top 5 improvement recommendations by impact
7. Action Plan: Generate specific, actionable improvement steps

Content Types

This skill supports the following content types, each with different dimension weights:

• - Product Review
• How-To Guide
• Comparison Review
• Landing Page
• Blog Post
• FAQ
• Alternative Recommendation
• Best Recommendation
• User Review

Usage

Basic Audit

CODEBLOCK0

CODEBLOCK1

Specify Content Type

CODEBLOCK2

CODEBLOCK3

Comparison Audit

CODEBLOCK4

Data Input Requirements

Manual Data Input (Currently recommended):

Request users to provide:

1. 1. Content text, URL, or file path
2. Content type (if cannot auto-detect): Product Review, How-To Guide, Comparison Review, Landing Page, Blog Post, FAQ, Alternative Recommendation, Best Recommendation, User Review
3. Optional: Competitor content for comparative assessment

Note: Explicitly mark in the output which items cannot be fully evaluated due to lack of access (e.g., backlink data, Schema markup, site-level signals).

Execution Steps

When users request content quality audit, follow these steps:

Step 1: Audit Preparation

CODEBLOCK5

If any veto item is triggered, prominently mark it at the top of the report and recommend immediate action before continuing with the full audit.

Step 2: CORE Audit (40 Items)

Evaluate each item according to standards in references/core-eeat-benchmark.md.

Score each item:

• - Pass = 10 points (Fully meets standard)
• Partial = 5 points (Partially meets standard)
• Fail = 0 points (Does not meet standard)

CODEBLOCK6

Evaluate O (Organization), R (Referenceability), and E (Exclusivity) in the same table format, 10 items per dimension.

Step 3: EEAT Audit (40 Items)

CODEBLOCK7

Evaluate Ept (Expertise), A (Authority), and T (Trust) in the same table format, 10 items per dimension.

For detailed 80-item ID lookup table and site-level item handling instructions, see references/item-reference.md.

Step 4: Scoring and Reporting

Calculate scores and generate final report:

CODEBLOCK8

Validation Checkpoints

Input Validation

• - [ ] Content source identified (text, URL, or file path)
• [ ] Content type confirmed (auto-detected or user-specified)
• [ ] Content sufficient for meaningful audit (≥300 words)
• [ ] If comparative audit, competitor content also provided

Output Validation

• - [ ] All 80 items scored (or marked N/A with reason)
• [ ] All 8 dimension scores calculated correctly
• [ ] Weighted total score matches content type weight configuration
• [ ] Veto items checked and marked if triggered
• [ ] Top 5 improvements sorted by weighted impact, not arbitrary order
• [ ] Each recommendation specific and actionable (not generic)
• [ ] Action plan includes specific steps and investment estimates

Success Points

1. 1. Start with Veto Items — T04, C01, R10 are one-vote veto items; they affect overall evaluation regardless of total score

1. 2. Focus on High-Weight Dimensions — Different content types prioritize different dimensions

1. 3. GEO-First Items are Critical for AI Visibility — If goal is AI citation, prioritize items marked with GEO 🎯

1. 4. Some EEAT Items Require Site-Level Data — Don't penalize content for things only observable at site level (backlinks, brand recognition)

1. 5. Use Weighted Scores, Not Just Raw Averages — Product reviews with strong exclusivity are more important than strong authority

1. 6. Re-Audit After Improvements — Run again to verify score improvements and catch regressions

Terminology

CORE (Content Quality)

• - C (Contextual Clarity): Contextual Clarity — Whether content is clear, accurate, and directly answers user questions
• O (Organization): Organization — Whether content has good structure, hierarchy, and navigation
• R (Referenceability): Referenceability — Whether content has sufficient data, evidence, and citations
• E (Exclusivity): Exclusivity — Whether content offers unique insights, data, and perspectives

EEAT (Source Credibility)

• - Exp (Experience): Experience — Whether author demonstrates actual usage experience
• Ept (Expertise): Expertise — Whether author demonstrates professional knowledge and skills
• A (Authority): Authority — Whether content source possesses authority and industry status
• T (Trust): Trust — Whether content is trustworthy

GEO (Generative Engine Optimization)

• - Content optimization for AI search engines (e.g., Google SGE, Bing Chat)
• Emphasizes direct answer capability, referenceability, and exclusivity

Reference Documents

• - references/core-eeat-benchmark.md — Complete 80-item benchmark with dimension definitions, scoring standards, and GEO-First item markers
• references/item-reference.md — Compact lookup table for all 80 item IDs + site-level item handling instructions + sample scored report

Notes

• - Read reference documents only when needed to maintain context conciseness
• When operations are fragile or require strong consistency, prioritize script execution with result validation
• Fully leverage the agent's language understanding and generation capabilities; avoid writing scripts for simple tasks
• This skill primarily serves Chinese users but retains industry terminology (CORE, EEAT, GEO) for alignment with international standards