An advanced Generative Engine Optimization (GEO) skill to audit domain citations across major LLM search engines and optimize pages for AI crawlers.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install gr-geo-cite
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install gr-geo-cite 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 gr-geo-cite?
gr-geo-cite is a technical Generative Engine Optimization (GEO) skill designed to help developers, publishers, and founders capture valuable search traffic in the LLM era. As traditional search engine algorithms give way to AI answer engines, driving traffic to your site increasingly requires models like Claude, ChatGPT, Perplexity, and Gemini to actively mention and cite your domain. This tool automates the process of tracking search phrases and analyzing whether your target pages are being extracted and referenced.
By implementing this specialized package from the Openclaw Skills index, developers can address a critical vulnerability: zero-citation visibility. The skill uses a structured 'GEO Trinity' methodology consisting of highly optimized /llms.txt maps, FAQ JSON-LD schema layouts, and high-density 'Citable Statistics' tables placed directly in top articles. Together, these files and formatting principles turn standard web content into highly readable feeds for AI crawlers, helping authors boost citation counts systematically.
gr-geo-cite Use Cases
- AI Search Citation Audits: Conduct automated or manual weekly sweeps across Claude, ChatGPT, Perplexity, and Gemini to trace domain references.
- Citability Pre-Publishing Assessment: Scan local drafts and URLs using a custom scoring matrix to identify optimization opportunities prior to publication.
- Automated llms.txt Management: Generate structured
/llms.txtdirectories automatically mapping top-performing posts and key statistics. - AI Crawler Optimization: Inject specialized 'Citable Statistics' blocks containing highly verifiable numbers and source links within the first 1000 words of web pages.
- Conversational Search Benchmarking: Keep record of search context, sentiment, and rankings compared to direct competitors.
How gr-geo-cite Works
- Establish Queries: Setup a static list of highly specific target queries reflecting real-world user intent (e.g., 'What is the best Product Hunt launch playbook for 2026?').
- Run Model Evaluations: Execute the queries across four primary model APIs (Anthropic, OpenAI, Perplexity, and Google AI Studio).
- Regular Expression Parsing: Scan response bodies for target domains and capture positive, neutral, or negative citation contexts.
- Structured Logging: Write telemetry data directly into a local, appending
data/geo-citations.jsonldatabase for downstream analysis. - Citability Scoring: Run
scripts/citability-scorer.pyover target HTML pages to compute local citability grades (0-100) on structured readability and statistical density. - Build/Release Cascades: Trigger schema injection during blog generation steps and update
/llms.txtdynamically to keep scrapers informed.
gr-geo-cite Setup
Set up your environment by exporting your secret keys to access conversational API services:
export ANTHROPIC_API_KEY="your_anthropic_key"
export OPENAI_API_KEY="your_openai_key"
export PERPLEXITY_API_KEY="your_perplexity_key"
export GEMINI_API_KEY="your_gemini_key"
Run the built-in citability scorer to audit live URLs or local files for search crawler friendliness:
# Audit a live page for LLM citability
python3 scripts/citability-scorer.py https://gingiris.tools/best-social-listening-tools-startups-2026
# Audit a local HTML draft before deployment
python3 scripts/citability-scorer.py --file local-draft.html
Run your weekly automated tracking pipeline to log AI citations and output performance reports:
python3 scripts/weekly-cite-check.py
gr-geo-cite Data Schema & Taxonomy
Citations Ledger (data/geo-citations.jsonl)
Every run of the citation check appends JSON-lines tracking events using the following structure:
| Field | Type | Description |
|---|---|---|
date |
String | ISO-8601 date signature of the execution |
ai |
String | The model audited (e.g., claude, gpt, perplexity, gemini) |
query |
String | The exact prompt queried against the LLM |
cited_domains |
Array | List of unique URLs extracted from citations |
mention |
String | Citation outcome (e.g., "cited", "not_cited") |
context_snippet |
String | A brief slice of text surrounding the brand mention |
rank_in_list |
Integer | Position of the brand URL within the output citation index |
Citability Scorer Dimension Weights
The citability-scorer.py engine computes an aggregate score out of 100 based on five structural dimensions:
- Answer Block Quality (30%): Evaluates explicit definition match patterns, physical answer positioning, and direct question-headings.
- Self-Containment (25%): Validates an optimal passage length (134-167 words), low pronoun density, and at least 3 proper nouns.
- Structural Readability (20%): Checks for sentence lengths of 10-20 words, clean markdown list structures, and logical paragraph breaks.
- Statistical Density (15%): Assesses frequency of hard data points (percentages, currency, measurement units, and dates).
- Uniqueness Signals (10%): Scans for original-research indicators, case study identifiers, and specific product names.
gr-geo-cite Advanced Features
- Zero-Dependency Scorer: Features a lightning-fast, pure standard library
citability-scorer.pymodule adapted from advanced research guidelines, requiring zero external requests or parsing libraries. - Automated llms.txt Compilation: Automatically aggregates metadata, top-performing pages, and citable metrics into a clean
/llms.txtroute for LLM crawlers. - Trigger Cascades: Integrates seamlessly with static site generation tasks (like
gr-blog-postdeployments) to automatically prompt writers to insert Citable Statistics tables. - CI/CD Quality Gates: Easily configures into deployment pipelines to block website building tasks if the computed citability index falls below acceptable score ranges (e.g., < 50/100).
SKILL.md
Loading
Related Openclaw Skills
gr-social-distill
An AI-powered content distribution agent that repurposes blog posts into highly optimized X threads, LinkedIn carousels, Xiaohongshu notes, and dev.to/Zenn articles.
gr-seo-patrol
An automated daily SEO and GEO patrol agent skill that tracks SERP keyword rankings, verifies site health indexes, and automatically repairs canonical links and internal metadata.
SEO/GEO Operations Agent
An autonomous SEO agent SOP designed to execute daily site audits, track keyword rankings, automate Schema markups, and run GEO workflows on autopilot.
Product Dev Ops Playbook
A comprehensive standard operating procedure skill that aligns product, engineering, and operations teams using a unified backlog and structured 10-day sprint cadences.
Competitor Research & Growth Flywheel Playbook
An AI-driven strategic SOP for reverse-engineering competitor growth flywheels, website milestones, social media propagation chains, and pricing structures.
gr-competitor
An advanced competitive intelligence skill that orchestrates high-speed, parallel web extraction to perform deep audits of competitor landing pages, pricing tiers, and content strategies.
Featured*
