Georithm

Georithm is a Claude Code skill set for auditing AI discoverability. It inspects how a site is exposed to AI crawlers, how readable its content is for answer extraction, how well its entities are described, and how ready it is for platform-specific surfaces such as ChatGPT, Perplexity, Gemini, Bing Copilot, and Google AI Overviews.

The project is designed for practical workflows: run an audit, identify the blockers, generate a structured report, and apply the fixes. It is built around reproducible checks rather than vague scoring.

---

Quick Start

macOS / Linux

curl -fsSL https://raw.githubusercontent.com/chefallan/georithm/main/install.sh | bash

Windows (Git Bash)

curl -fsSL https://raw.githubusercontent.com/chefallan/georithm/main/install-win.sh | bash

Manual Install

git clone https://github.com/chefallan/georithm.git
cd georithm
./install.sh

Requirements:

• Python 3.8+
• Claude Code CLI
• Git
• Optional: Playwright for screenshot and browser-assisted checks

---

Commands

Command	Purpose
/georithm audit <url>	Full multi-skill audit for a site or landing page
/georithm quick <url>	Fast snapshot of AI visibility blockers
/georithm citability <url>	Passage-level answer extraction and citation readiness
/georithm crawlers <url>	robots.txt and AI bot access analysis
/georithm llmstxt <url>	Validate or generate llms.txt
/georithm brands <url>	Check entity and brand presence across major sources
/georithm platforms <url>	Platform-by-platform readiness review
/georithm schema <url>	Structured data detection and JSON-LD recommendations
/georithm technical <url>	Technical audit focused on rendering, crawlability, and performance
/georithm content <url>	E-E-A-T and content quality review
/georithm report <url>	Consolidated markdown report
/georithm report-pdf	PDF export from collected audit data
/georithm prospect ...	Prospect tracking and CRM-lite workflows
/georithm proposal ...	Proposal generation from audit output
/georithm compare ...	Progress and delta reporting across audits

---

What Georithm Checks

AI crawler access

Georithm inspects robots.txt, page-level directives, and related technical signals to show which AI bots can reach your content and which are blocked.

Content extractability

It scores whether a page answers questions clearly enough for AI systems to quote or summarize it with confidence.

Entity and schema quality

It reviews sameAs links, organization and author markup, and related entity data so that a site is easier to resolve as a trusted source.

Platform-specific readiness

It separates platform concerns instead of treating AI visibility as one generic score. The same page may perform differently across Google, Bing, ChatGPT-style search, and community-heavy citation systems.

Reporting

It can produce markdown and PDF output suitable for internal audits, client delivery, or recurring monitoring.

---

Repository Layout

georithm/
├── georithm/                     # Main orchestration skill
├── skills/                       # Task-specific sub-skills
├── agents/                       # Parallel analysis agents
├── scripts/                      # Python utilities used by the skills
├── schema/                       # JSON-LD templates and examples
├── examples/                     # Example audit and proposal output
├── tests/                        # Automated tests for helper scripts
├── install.sh                    # Installer for macOS/Linux
├── install-win.sh                # Installer for Git Bash on Windows
└── uninstall.sh                  # Uninstaller

---

Runtime Data

Prospect and reporting workflows store data outside the repository in:

~/.georithm-prospects/
├── prospects.json
├── audits/
├── proposals/
└── reports/

The uninstaller does not remove that directory automatically.

---

Audit Model

The default full audit combines the following areas:

Area	Weight
AI Citability & Visibility	25%
Brand / Entity Signals	20%
Content Quality & E-E-A-T	20%
Technical Foundations	15%
Structured Data	10%
Platform-Specific Readiness	10%

These weights are intended as a working model for prioritization, not as a claim of industry-standard certification.

---

Common Uses

• Agency audits and client reporting
• Internal content QA for AI discoverability
• Migration reviews for rendering or schema regressions
• Local business visibility checks
• SaaS and documentation site audits
• E-commerce entity and product markup reviews

---

Uninstall

./uninstall.sh

Or manually:

rm -rf ~/.claude/skills/georithm ~/.claude/skills/georithm-* ~/.claude/agents/georithm-*.md

---

Contributing

Issues and pull requests are welcome. If you are changing scoring logic or output formats, keep the change narrow and document the reasoning in the relevant skill file.

---

License

Released under the MIT License. See the LICENSE file for the full text.