Causabi developer reference
causabi-geo is an open-source (MIT) CLI that audits a page for AI-crawler readiness and generates ready-to-deploy fix files — locally, no account needed. Continuous monitoring and multi-page audits run on causabi.com.
Quickstart
pip install causabi-geo geo-optimizer analyze https://yourdomain.com
causabi-geo works as an alias for geo-optimizer since v0.2.1.
CLI commands
# Analyze — AI Readiness Score 0-100 + issue list geo-optimizer analyze https://yourdomain.com geo-optimizer analyze https://yourdomain.com --json # Score — alias for analyze geo-optimizer score https://yourdomain.com # Fix — generates ready-to-deploy files geo-optimizer fix https://yourdomain.com geo-optimizer fix https://yourdomain.com --output ./geo-fixes # Fix with AI-generated FAQ (requires a Gemini API key) geo-optimizer fix https://yourdomain.com --api-key YOUR_GEMINI_KEY
fix writes robots.txt (patched to allow AI crawlers), schema.json, llms.txt, faq-schema.json (with --api-key) and HOW-TO-APPLY.md.
What the score measures
Six categories: robots.txt access for AI crawlers (20), Schema.org completeness (25), FAQ schema (20), content depth (15), brand / NAP signals (10), freshness (10). The score measures crawlability and machine-readability — it does not guarantee citations.
MCP server
The MCP server lets Claude Code, Claude Desktop, Cursor and any MCP-compatible client check GEO scores and generate fixes from your editor. Runs locally over stdio.
# Claude Code claude mcp add geo-optimizer -- npx -y geo-optimizer-mcp # Then ask: # > Check the GEO score for example.com
Config-file setup for other clients: geo-optimizer-mcp on GitHub.
Hosted API
A public score endpoint is available without a key:
curl https://ai.causabi.com/api/score/yourdomain.com
Authenticated endpoints (full audits, monitoring, file generation) require an API key — create one in the dashboard settings. See /for-developers for integration examples.