GEO Audit Skill for Openclaw

A professional Generative Engine Optimization (GEO) diagnostic tool that identifies why AI systems fail to discover or cite your website.

enzyme2013
v1.2.0
Apr 6, 2026
1
845
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install geo-audit

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install geo-audit using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is GEO Audit Skill?

The GEO Audit skill is a sophisticated diagnostic agent designed to help developers and marketers navigate the era of AI-driven search. It performs a comprehensive analysis of a website across four critical dimensions: Technical Accessibility, Content Citability, Structured Data, and Brand Signals. By simulating how modern LLMs and generative engines perceive a site, it provides a data-backed assessment of your digital footprint within the framework of Openclaw Skills.

Built on a research-backed 3-layer model, this skill goes beyond traditional SEO to focus on Generative Engine Optimization (GEO). It evaluates how likely your content is to be used as a primary source by platforms like ChatGPT, Perplexity, and Claude, ensuring your brand remains authoritative and discoverable as user behavior shifts from search engines to AI agents.

GEO Audit Skill Use Cases

  • Investigating why a website is not being cited or recommended by ChatGPT, Claude, or Perplexity.
  • Performing a technical audit to ensure AI crawlers are not blocked by robots.txt or rendering issues.
  • Evaluating the citability of content to increase the chances of being featured in AI-generated answers.
  • Generating structured data templates (JSON-LD) specifically optimized for AI entity recognition.

How GEO Audit Skill Works

  1. The skill starts by validating the target URL and detecting the business type (SaaS, E-commerce, Agency, etc.) to tailor the scoring weights.
  2. It fetches the homepage and analyzes robots.txt and sitemaps to identify up to 10 key pages for a representative sample.
  3. Four specialized subagents (Technical, Citability, Schema, and Brand) are launched in parallel to analyze different aspects of the site.
  4. Each subagent returns a score based on a granular rubric, which is then aggregated into a composite GEO Score (0-100).
  5. The skill generates a prioritized fix plan, ranging from critical technical blocks to low-priority content tweaks.

GEO Audit Skill Setup

To integrate this diagnostic capability into your workflow using Openclaw Skills, ensure you have the subagent instruction files located in the references/agents/ directory.

# Navigate to your skills directory
cd path/to/your/agent/skills

# Ensure the geo-audit files are present
ls geo-audit/references/agents/
# Expected: geo-technical.md, geo-citability.md, geo-schema.md, geo-brand.md

Once files are in place, you can trigger the audit by providing any URL to the agent and asking for a GEO audit.

GEO Audit Skill Data Schema & Taxonomy

The skill produces a detailed Markdown report and maintains a machine-readable metadata block for historical tracking. The data is organized as follows:

Dimension Weight Primary Focus
Technical 20% AI crawler access, rendering, and meta header signals.
Citability 35% Answer block quality, statistical density, and expertise signals.
Schema 20% Core identity schema (JSON-LD) and content-specific markup.
Brand 25% Entity recognition, community signals, and third-party presence.

All reports include a GEO-AUDIT-META block containing the domain, date, business type, and individual dimension scores.

GEO Audit Skill Advanced Features

  • Parallel Multi-Agent Execution: Dispatches four independent subagents simultaneously to minimize processing time.
  • Security Sandboxing: Uses untrusted-content wrappers for all external data to prevent prompt injection attacks from malicious websites.
  • Platform-Specific Biases: Provides customized advice based on the unique citation behaviors of ChatGPT, Gemini, and Claude.
  • Automated JSON-LD Generation: Creates ready-to-use schema templates based on detected missing data points.
  • Priority-Sorted Roadmap: Automatically categorizes issues by point impact to help users focus on high-ROI fixes.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*