geo

GEO Visibility Tracking

Track how your brand appears across AI search engines — ChatGPT, Perplexity, Gemini, Claude, and Grok. GEO fans out a prompt to multiple LLMs simultaneously and analyzes mentions, position, sentiment, citations, competitor share of voice, and more. Pay-per-use through the Obul proxy.

When to Use

• User wants to check if/how a brand is mentioned by AI search engines
• User wants to compare brand visibility against competitors across LLMs
• User needs sentiment analysis of how AI models talk about a brand
• User wants prompt suggestions that would lead to brand mentions
• User wants to analyze citation URLs that AI models reference
• User is doing GEO/AIO (AI optimization) research or auditing

Endpoints

Visibility Check

• URL: https://geo.x402endpoints.com/v1/visibility/check
• Method: POST
• Pricing: $0.03 per request

Fan out a prompt to multiple LLMs and get a full visibility report: mentions, paragraph position, sentiment, citations with linked URLs, competitor analysis, and share of voice.

Request:

obulx -X POST -H "Content-Type: application/json" \
  -d '{
    "prompt": "What is the best project management tool for startups?",
    "brand": "Linear",
    "aliases": ["linear.app"],
    "competitors": ["Jira", "Asana", "Monday"],
    "models": ["perplexity", "openai", "gemini", "claude", "grok"]
  }' \
  "https://geo.x402endpoints.com/v1/visibility/check"

Parameters:

Parameter	Type	Default	Description
prompt	string	required	The search prompt to test
brand	string	required	Brand name to track
aliases	array	[]	Alternative brand names or domains
competitors	array	[]	Competitor names to track
models	array	all 5	Which LLMs to query: "perplexity", "openai", "gemini", "claude", "grok"

Response:

{
  "brand": "Linear",
  "prompt": "What is the best project management tool for startups?",
  "timestamp": "2026-03-02T12:00:00.000Z",
  "models": [
    {
      "modelId": "perplexity",
      "modelName": "Perplexity Sonar",
      "mention": {
        "mentioned": true,
        "position": 0.0,
        "paragraphPosition": 1,
        "positionLabel": "first_paragraph",
        "count": 3,
        "matchedTerm": "Linear"
      },
      "sentiment": {
        "label": "positive",
        "confidence": 0.6
      },
      "citations": {
        "markers": ["[1]", "[2]"],
        "urls": ["https://linear.app", "https://example.com/review"],
        "linked": [
          { "marker": "[1]", "url": "https://linear.app" },
          { "marker": "[2]", "url": "https://example.com/review" }
        ]
      },
      "competitors": [
        { "name": "Jira", "mentioned": true, "count": 2 },
        { "name": "Asana", "mentioned": true, "count": 1 },
        { "name": "Monday", "mentioned": false, "count": 0 }
      ],
      "latencyMs": 1200,
      "rawTextLength": 850
    }
  ],
  "errors": [],
  "summary": {
    "mentionRate": 0.8,
    "avgPosition": 0.1,
    "dominantSentiment": "positive",
    "totalModelsQueried": 5,
    "totalModelsSucceeded": 5,
    "shareOfVoice": {
      "brand": 0.5,
      "competitors": [
        { "name": "Jira", "share": 0.3 },
        { "name": "Asana", "share": 0.15 },
        { "name": "Monday", "share": 0.05 }
      ]
    }
  }
}

Prompt Suggest

• URL: https://geo.x402endpoints.com/v1/prompts/suggest
• Method: POST
• Pricing: $0.01 per request

Generate AI-powered prompt suggestions that would lead to brand mentions. Useful for content strategy and GEO optimization.

Request:

obulx -X POST -H "Content-Type: application/json" \
  -d '{
    "url": "https://linear.app",
    "brand": "Linear",
    "stage": "consideration"
  }' \
  "https://geo.x402endpoints.com/v1/prompts/suggest"

Parameters:

Parameter	Type	Default	Description
url	string	required	Brand website URL
brand	string	required	Brand name
stage	string	"awareness"	Funnel stage: "awareness", "consideration", "decision"

Response:

{
  "brand": "Linear",
  "url": "https://linear.app",
  "prompts": [
    {
      "prompt": "What are the fastest issue trackers for engineering teams?",
      "stage": "consideration",
      "rationale": "Targets Linear's core differentiator of speed and developer experience"
    }
  ]
}

Citation Analyze

• URL: https://geo.x402endpoints.com/v1/citations/analyze
• Method: POST
• Pricing: $0.01 per request

Fetch and analyze citation URLs for content quality metrics — word count, entity density, structured data, and freshness.

Request:

obulx -X POST -H "Content-Type: application/json" \
  -d '{
    "urls": [
      "https://linear.app/blog/post-1",
      "https://example.com/linear-review"
    ]
  }' \
  "https://geo.x402endpoints.com/v1/citations/analyze"

Parameters:

Parameter	Type	Default	Description
urls	array	required	Citation URLs to analyze (1-10)

Response:

{
  "citations": [
    {
      "url": "https://linear.app/blog/post-1",
      "title": "Why Linear is Fast",
      "wordCount": 1200,
      "entityDensity": 0.15,
      "hasStructuredData": true,
      "freshness": "2026-02-15"
    }
  ]
}

Endpoint Pricing Reference

Endpoint	Price	Purpose
/v1/visibility/check	$0.03	Full visibility analysis across up to 5 LLMs
/v1/prompts/suggest	$0.01	AI-generated prompt suggestions for brand visibility
/v1/citations/analyze	$0.01	Analyze citation URL quality metrics
/health	$0.00	Health check
/pricing	$0.00	Get per-endpoint pricing
/openapi.json	$0.00	OpenAPI 3.1 specification

Best Practices

• Start with a visibility check using all 5 models to get a baseline, then narrow to specific models for follow-up
• Use aliases for brands with common abbreviations (e.g., "Amazon Web Services" + "AWS")
• Include 2-4 direct competitors to get meaningful share of voice data
• Use the positionLabel field to quickly assess brand prominence ("first_paragraph" is ideal)
• Chain visibility check → prompt suggest → re-check to measure optimization impact
• The linked array in citations tells you which URLs are actually cited by number — more actionable than raw URL lists

Error Handling

Error	Cause	Solution
400	Missing required fields (prompt, brand, url, urls)	Check request body
402	No payment or insufficient balance	Verify your account has sufficient balance at my.obul.ai. Run obulx login if not authenticated.
429	Rate limit exceeded	Wait and retry after Retry-After header value
502	Upstream LLM failed	Retry; check errors array for per-model failures
503	Service misconfigured	Contact service operator