LSD
APIEndpoint 19 of 40AI Optimization / Sources / LLM Authority

AI Top Sources API

Which domains do LLMs trust for your keywords.

In the LLM era, ranking means being one of 3–6 sources an AI system cites in its answer. When someone asks ChatGPT 'best plumber in Austin,' the system synthesizes an answer and cites 4–6 domains. Getting onto that list is part of how you drive real traffic in 2026. The dashboard era never built tools to see which domains win that citation race. This endpoint returns the domains ChatGPT, Claude, Gemini, and Perplexity cite most often for your keywords — ranked by mention frequency. Use it to identify link-building targets, understand your industry's citation authority hierarchy, and spot where your own domain ranks against competitors.

Start Free — 50 CreditsView Docs

POST /v1/ai/top-sources · 5 credits / call

POST /v1/ai/top-sources4 of 18
Domains LLMs cite for topic
ChatGPTcited 12× / week

ABC Plumbing is highly rated in Austin with 4.8 stars and excellent same-day service.

Google AI Overview1 of 3 cited

Top-rated Austin plumbers include ABC Plumbing for emergency service.

Claudecited 8× / week

For licensed plumbers in Austin, ABC Plumbing has strong reviews.

Perplexitycited 5× / week

ABC Plumbing — 24/7 emergency response, 500+ reviews.

▌ Ask your agent

These prompts identify your link-building priorities in LLM eyes.

Connect Local SEO Data as an MCP server once (60 seconds, below). Then your agent discovers which domains LLMs trust for your market. Replace bracketed keywords with your own.

Link-building target discovery

Pull the top 10 domains LLMs cite most for [your industry keyword]. Filter to domains with domain authority > 50. These are your link-building whitelist — get cited there and you'll gain authority in AI eyes.

Citation authority audit

Run AI Top Sources for [your keyword] across ChatGPT, Claude, and Gemini. Show me the rank of [mycompany.com] vs [competitor1.com] and [competitor2.com]. Which competitor appears in the top 5 most-cited domains?

Market authority landscape

Pull the top 20 most-cited domains for [industry keyword] and [related keyword]. Are the same domains dominant across both? Which new sources are emerging in the citation landscape? This tells me where the LLM authority bar is for our market.

Competitive citation gaps

Compare top-cited domains for [keyword] across all four LLMs. Which domains appear in ChatGPT but not Gemini? Which ones are cited by every LLM? Find the gaps where we can shift domain preferences by improving our positioning.

▌ Real response

What you get back

Live response from POST /v1/ai/top-sources for 'personal injury lawyer Denver' across ChatGPT and Google AI.

response · application/json~5s · 5 credits
{
  "status": "success",
  "credits_used": 5,
  "data": {
    "keyword": "personal injury lawyer Denver",
    "location": "Denver, Colorado",
    "ai_search_volume": 2100,
    "top_domains": [
      {
        "domain": "nolo.com",
        "mentions": 42,
        "ai_search_volume": 2100,
        "citation_count": 18
      },
      {
        "domain": "avvo.com",
        "mentions": 38,
        "ai_search_volume": 2100,
        "citation_count": 16
      },
      {
        "domain": "justia.com",
        "mentions": 35,
        "ai_search_volume": 2100,
        "citation_count": 15
      },
      {
        "domain": "martindale.com",
        "mentions": 28,
        "ai_search_volume": 2100,
        "citation_count": 12
      },
      {
        "domain": "denverinjurylaw.com",
        "mentions": 12,
        "ai_search_volume": 2100,
        "citation_count": 5
      }
    ]
  }
}
▌ Returns

The citation authority map for any keyword

Domain citation frequency

How many times each domain is cited

Not all top-ranked domains are cited equally. We capture how many times each domain appears in LLM responses for your keyword, giving you the pecking order of authority in AI eyes.

Per-domain mention context

Citation volume and consistency

See which domains appear in ChatGPT answers, which in Claude, which in Gemini. Each LLM has different training data — a domain trusted by Claude may not appear in Perplexity responses. Track per-LLM separately.

Link-building whitelist

High-authority citation targets

Domains cited frequently by LLMs are flagged as high-authority in AI's worldview. Getting a link or citation on one of these domains signals authority to AI systems — a more direct ROI than traditional backlink building.

Your domain's citation rank

Where you stand against competitors

See your position in the top-cited domains list. If you rank #15 on a keyword, you know exactly which 14 competitors or authorities outrank you in AI citations. This is your visibility gap.

▌ Built for

What AI-native operators ship with this

Link-building strategy prioritization

Traditional SEO link builders target high-domain-authority sites. In 2026, LLM visibility operators target high-citation-frequency sites. Pull AI Top Sources for your core keywords and build a whitelist of domains that appear in 30+ LLM responses. These are your link targets.

→ For link builders

Citation gap analysis

Run AI Top Sources for your keyword and pull the same keyword for competitors. See which domains cite your competitors but not you. These are the citation gaps in your market positioning. Close them via PR, content strategy, or partnerships.

→ Citation Audit API

GEO and AEO strategy

Generative Engine Optimization (GEO) and Answer Engine Optimization (AEO) require understanding which domains the engines trust. This endpoint is the data layer for GEO strategy — tell your content team which authoritative sources to cite, which domains to target for mentions, and how to position against competitors in AI eyes.

→ For content strategists

Authority repositioning

If your domain doesn't appear in the top 10 most-cited sources for your keywords, you need to reposition. Use this endpoint weekly to track whether you're climbing the citation ranks. Pair with AI Mentions to understand which queries mention you and which don't.

→ AI Mentions API
▌ vs. the alternatives

Why not track top sources manually or via traditional tools?

Checking which domains ChatGPT, Claude, Gemini cites for your keywords by hand is tedious and unrepeatable. Competitors charge $200–$2000 per month for source tracking. We charge $0.025 per call, unlimited keywords. Here's how the options stack up.

ApproachCost per source checkLLMs trackedLink-building dataAgent-ready
Manual checking (ChatGPT individually)$0 but ~15 min labor per keywordWhatever you check manuallyYou see the sourcesNo
Ahrefs / Semrush backlink tools$99–$999/mo (~$3–30 per query)Google ranking data onlyTraditional backlink targetsLimited
Profound source tracking$499–$2000/mo (~$16–65 per query)10+ platformsCitations, sources, authorityDashboard only
AthenaHQ$300–$1000/mo (~$10–33 per query)ChatGPT, Claude, Gemini, PerplexityCitation framing, narrative positioningDashboards + limited API
Otterly.ai source tracking$29–$99/mo (~$0.97–3.30 per query)4 LLMsDomain citations onlyUI export
BrightEdge source insights$500+/mo + seatsLimited LLM coverageEnterprise onlyDashboard
Local SEO Data AI Top Sources API$0.025 per callChatGPT, Claude, Gemini, PerplexityDomain citations, mention frequencyNative MCP, agent-first
▌ Connect in 60 seconds

Use it from your agent

Two integration surfaces: MCP for clients that speak MCP, REST API for everything else.

Direct MCP integration

Drop-in support in Claude Desktop, OpenClaw, Hermes Agent, and any MCP-aware client.

Add to your client's MCP config (e.g. claude_desktop_config.json):

{
  "mcpServers": {
    "localseodata": {
      "url": "https://mcp.localseodata.com",
      "headers": {
        "Authorization": "Bearer sk_live_..."
      }
    }
  }
}

REST API

For Perplexity Computer, ChatGPT Custom GPTs, custom agents, and any platform that calls REST endpoints directly.

Base URL:

api.localseodata.com

See the docs for endpoint reference and auth.

▌ Quickstart

Your first call in three lines

Core parameters: `keyword` (the topic to analyze) and optional `location` (e.g. 'Denver, CO', defaults to United States). You can specify which LLM platforms to query — default is both ChatGPT and Google (AI Overview). The `limit` parameter caps results (default 10, max 20). One call = 5 credits ($0.025). Returns the top-cited domains for your keyword, ranked by mention frequency.

terminal · curl
POST /v1/ai/top-sources
curl -X POST https://api.localseodata.com/v1/ai/top-sources \
  -H "Authorization: Bearer sk_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "keyword": "personal injury lawyer Denver",
    "location": "Denver, Colorado",
    "platforms": ["chat_gpt", "claude", "gemini", "perplexity"],
    "limit": 10
  }'
Full endpoint reference → · Get an API key →
▌ Pricing for this endpoint

$0.025 per top-sources query

Pay-as-you-go starts at $5. No monthly minimums. No per-seat licenses. No subscription required. Funds never expire.

Free tier
50
credits on signup (10 queries)
Starter · $5
200
credits, no expiration
Per-call cost
$0.025
per source query
▌ FAQ

Common questions

What is the AI Top Sources API?+
A REST endpoint that returns the domains LLMs cite most frequently for your keyword. For each domain, you get mention count, citation frequency, and which LLMs (ChatGPT, Claude, Gemini, Perplexity) cite it. The endpoint is POST /v1/ai/top-sources; one call costs 5 credits (~$0.025). This is the data layer underneath link-building strategy and generative engine optimization (GEO). If you want to know which domains ChatGPT trusts most for 'best pizza Brooklyn' or 'personal injury lawyer Denver,' this endpoint answers that.
How is this different from AI Visibility and AI Mentions?+
AI Visibility is a composite score (0–100) measuring your overall domain visibility across LLMs. AI Mentions returns the exact question-answer-context for every time your brand is mentioned in LLM responses. AI Top Sources returns the ranking of which *other* domains LLMs cite most for a keyword — the authority hierarchy. They're complementary: Visibility tells you 'how visible you are,' Mentions tells you 'where and how you appear,' Top Sources tells you 'which competitors and authorities are cited more often.' Many teams query all three: Top Sources to identify link targets, Visibility to track your position, Mentions to understand your framing.
Which LLMs do you track?+
ChatGPT (OpenAI), Claude (Anthropic), Gemini (Google), Perplexity (Perplexity AI), and Google AI Overviews. These are the major LLM surfaces driving real traffic in 2026. Each has distinct citation behavior — a domain trusted by Claude may not appear in Gemini responses. You can query all platforms in one call or filter to specific ones. If you need coverage beyond these five (Grok, DeepSeek, Copilot), let us know.
How should I use this for link-building?+
Link-building in the LLM era is about getting cited by high-authority sources *that LLMs also cite frequently*. Traditional SEO targets high-domain-authority sites. In 2026, you target high-citation-frequency sites — domains that LLMs mention repeatedly. Use this endpoint to build your link-building whitelist: pull top sources for your 5–10 core keywords, filter to domains with domain authority > 50, and prioritize getting mentioned on those sites. If you appear on a domain cited 30+ times by LLMs, your visibility in AI search jumps.
How fresh is the top-sources data?+
Each API call executes your keyword live across ChatGPT, Claude, Gemini, and Perplexity, so the top sources reflect current LLM training and behavior. LLM responses vary between calls, so we execute multiple times per keyword to stabilize the signal. Results are available within 5–10 seconds of the API call. For trend tracking, run this weekly to see if you're climbing or falling in the citation authority ranks.
Can I track top sources for competitor keywords?+
Yes. The API accepts any keyword — yours, your competitors', or generic market queries like 'best pizza Brooklyn' or 'emergency plumber Austin'. There's no ownership verification. This is what makes competitive intelligence work: pull top sources for your keyword and your competitor's keyword, see which domains dominate both, and understand the citation authority landscape in your market.
Where does this citation data come from?+
We operate first-party collection infrastructure — our agents run your keywords directly against ChatGPT API, Claude API, Gemini API, Perplexity API, and Google AI Overviews. We execute the queries, capture the full responses, extract which domains are cited, count citation frequency, and return the ranking. We do not rely on cached or third-party data — each call executes fresh. This is why you see current citation authority, not historical data.
How does this compare to Profound, AthenaHQ, Otterly, and Peec AI?+
Profound ($499–$2000/mo), AthenaHQ ($300–$1000/mo), Otterly ($29–$99/mo), and Peec AI ($99–$299/mo) all offer dashboards that track which domains are cited for keywords. Profound is enterprise-focused; AthenaHQ emphasizes narrative positioning; Otterly is budget-friendly. All are designed for humans logging in and reading reports. Our API is designed for agents — your Claude prompt calls this endpoint, gets JSON, and acts on it without a dashboard. Pricing-wise: Profound costs $16–65 per query annualized; AthenaHQ costs $10–33; Otterly costs $0.97–3.30; we charge $0.025 per call. If you're building agent workflows or running 100+ source checks per month, API pricing wins. If you want a CEO dashboard, those tools are better suited.
Can my AI agent use this directly?+
Yes, and this is exactly what we built for. Two integration paths: MCP: add Local SEO Data to your claude_desktop_config.json and your Claude agent calls this endpoint from any prompt. REST: any agent that can make HTTPS calls (ChatGPT Custom GPTs, Perplexity Computer, custom Python/Node agents) hits the API directly with your Bearer token. The agent receives structured JSON (domain, mentions, ai_search_volume) and can filter, rank, alert, or feed it into your link-building workflow. No integration code needed — just write the prompt.
What does it cost compared to competitors?+
We charge $0.025 per top-sources query (one keyword across selected LLMs). Profound costs $499–$2000/mo (~$16–65 per query annualized). AthenaHQ costs $300–$1000/mo (~$10–33 per query). Otterly costs $29–$99/mo (~$0.97–3.30 per query). Peec AI costs $99–$299/mo. If you run 100 source checks per month, we cost $2.50 total; Profound costs $500+. If you run 10 checks per month, we cost $0.25; Profound still costs $500. Pay-per-call eliminates the monthly minimum that subscription tools impose.
What changed in 2026 that made AI source tracking necessary?+
Three things: ChatGPT, Claude, Gemini, and Perplexity are now mainstream surfaces generating real user traffic. When someone asks these systems a question, they cite 3–6 sources in their answer. Getting onto that citation list is now part of traffic generation. Second, LLM citations are concentrated — the top 3–6 domains get the vast majority of citations for any keyword. In traditional Google search, rankings distribute across hundreds of results. In LLM search, citation authority is narrower, meaning fewer winners and higher stakes for those on the list. Third, MCP (Model Context Protocol) made it practical for agents to call specialized APIs without custom integration. The dashboard era was built for humans. The agent era needs APIs that return ranked domain data. AI Top Sources is a 2026 native category because citation authority is now a real traffic surface and agents need the data to identify link-building targets.
How does generative engine optimization (GEO) use this endpoint?+
Generative Engine Optimization (GEO) is the 2026 equivalent of traditional SEO optimization for LLM visibility. The first step of GEO is understanding which domains and content formats the engines trust. This endpoint tells you which domains LLMs cite most for your keywords. Your content team uses this data to: (1) identify which authoritative sources to cite in your own content (show AI you understand the domain authority hierarchy), (2) target co-citation opportunities (get mentioned alongside already-trusted sources), (3) reposition your content strategy toward the format and sources LLMs prefer. In short: use AI Top Sources to understand the authority landscape, then create content that positions you as an authority within it.
▌ Related endpoints

Often used in the same agent prompt

POST /v1/ai/visibility
AI Visibility

Your domain's overall visibility score across ChatGPT, Claude, Gemini, and Perplexity.

POST /v1/ai/mentions
AI Mentions

Every mention of your brand in LLM responses with question, answer, and sentiment context.

POST /v1/ai/top-pages
AI Top Pages

Specific pages (not just domains) that LLMs cite most. More granular than top sources.

POST /v1/ai/compare
AI Compare

Compare your domain vs competitors on AI visibility, mentions, and citations.

Find your link-building targets in LLM eyes.

50 free credits on signup. Your first top-sources query happens through your agent, not a dashboard.

Start FreeRead the docs
▌ MADE FOR THE NEW LOCAL SEO STACK