Skip to main content

Configuration

GEO Optimizer ships with pre-built AI context files for six platforms. Once loaded, your AI assistant understands GEO methodology and can run the scripts, interpret audit results, and guide content optimization -- all without you needing to explain the toolkit from scratch each time.

Overview​

PlatformFileChar LimitSetup Method
Claude Projectsclaude-project.md~11,700 charsWeb UI paste
ChatGPT Custom GPTchatgpt-custom-gpt.md8,000 charsGPT Builder
ChatGPT Instructionschatgpt-instructions.md1,500 charsSettings paste
Cursorcursor.mdcUnlimitedFile in .cursor/rules/
Windsurfwindsurf.mdUnlimitedFile in .windsurf/rules/
Kirokiro-steering.mdUnlimitedFile in .kiro/steering/

All context files live in the ai-context/ directory of your GEO Optimizer installation.

SKILL.md

The SKILL.md file at the repository root serves as a platform-agnostic index. It describes all available context files and their deployment paths. Some AI platforms that support project-level documentation will read this file automatically.

Claude Projects​

Claude Projects supports the most comprehensive context, with no practical character limit for project instructions.

Setup​

  1. Open Claude and navigate to your project
  2. Click the project name to open settings
  3. Scroll to "Custom Instructions" (or "Project Instructions")
  4. Open ai-context/claude-project.md in a text editor: 
    cat ~/geo-optimizer-skill/ai-context/claude-project.md
  5. Copy the entire content and paste it into the project instructions field
  6. Save

What It Includes​

The Claude context file is the most detailed, covering:

  • Full methodology overview with the 9 Princeton methods
  • Complete script reference (flags, output formats, examples)
  • Scoring algorithm breakdown
  • AI bot reference table
  • Workflow guidance for interpreting and acting on audit results

ChatGPT Custom GPT​

For building a dedicated GEO Optimizer GPT with a richer context than standard chat.

Setup​

  1. Go to ChatGPT and open the GPT Builder (Explore GPTs > Create)
  2. In the Configure tab, find the Instructions field
  3. Open ai-context/chatgpt-custom-gpt.md: 
    cat ~/geo-optimizer-skill/ai-context/chatgpt-custom-gpt.md
  4. Paste the content into the Instructions field
  5. Name your GPT (e.g., "GEO Optimizer") and save
Character Limit

The Custom GPT instructions field has an 8,000 character limit. The chatgpt-custom-gpt.md file is optimized to fit within this constraint while preserving the essential methodology, script reference, and scoring details.

ChatGPT Custom Instructions​

For users who want GEO knowledge in their regular ChatGPT conversations without creating a dedicated GPT.

Setup​

  1. Open ChatGPT > Settings > Personalization > Custom Instructions
  2. Open ai-context/chatgpt-instructions.md: 
    cat ~/geo-optimizer-skill/ai-context/chatgpt-instructions.md
  3. Paste into the "How would you like ChatGPT to respond?" field
  4. Save
Compression

This file is heavily compressed to fit within the 1,500 character limit. It covers core concepts and script names but omits detailed flag references. For full context, use the Custom GPT approach instead.

Cursor​

Cursor reads rules files from .cursor/rules/ in your project directory.

Setup​

# From your project root
mkdir -p .cursor/rules
cp ~/geo-optimizer-skill/ai-context/cursor.mdc .cursor/rules/geo-optimizer.mdc

How It Works​

Cursor automatically loads all .mdc files from .cursor/rules/ as context for AI-assisted coding. With the GEO Optimizer rules loaded, Cursor's AI can:

  • Generate JSON-LD schemas matching your framework
  • Suggest robots.txt configurations for AI bots
  • Help write llms.txt content
  • Apply Princeton GEO methods to content
Multiple Rule Files

The .cursor/rules/ directory supports multiple files. Your existing rules will not be overwritten.

Windsurf​

Windsurf uses a similar rules directory structure.

Setup​

# From your project root
mkdir -p .windsurf/rules
cp ~/geo-optimizer-skill/ai-context/windsurf.md .windsurf/rules/geo-optimizer.md

Windsurf loads all files from .windsurf/rules/ as AI context automatically.

Kiro​

Kiro uses steering files to guide its AI behavior.

Setup​

# From your project root
mkdir -p .kiro/steering
cp ~/geo-optimizer-skill/ai-context/kiro-steering.md .kiro/steering/geo-optimizer.md

Kiro reads all markdown files from .kiro/steering/ and uses them to inform its responses.

Verifying the Setup​

After configuring your chosen platform, test with a simple prompt:

What is my GEO score for https://example.com?
Run a geo_audit and tell me what to fix.

If the context is loaded correctly, the AI should understand the GEO methodology, know the script names and flags, and provide structured guidance based on the audit results.

Updating Context Files​

When you update GEO Optimizer via update.sh, the context files in ai-context/ are updated automatically. However, you will need to re-paste updated content into web-based platforms (Claude Projects, ChatGPT) manually.

For file-based platforms (Cursor, Windsurf, Kiro), re-run the cp command:

# Cursor
cp ~/geo-optimizer-skill/ai-context/cursor.mdc .cursor/rules/geo-optimizer.mdc

# Windsurf
cp ~/geo-optimizer-skill/ai-context/windsurf.md .windsurf/rules/geo-optimizer.md

# Kiro
cp ~/geo-optimizer-skill/ai-context/kiro-steering.md .kiro/steering/geo-optimizer.md

Next: API Reference -- detailed flag reference, JSON output schema, and CI/CD integration.