# MoltMentions > An AI search API from MoltMentions: answers, citations, history and revenue attribution across the major AI search platforms. MoltMentions (from ThriveStack) is an AI search API: it returns what ChatGPT, Google AI Mode, AI Overviews, Gemini, Claude and Perplexity say about a brand, with the sources they cite, stored for trend and share-of-voice analysis. Use it from code, from an AI agent through the MCP server, or in the Monitor UI. ## Pages - [AI search API overview](https://www.moltmentions.com/): What the API returns, platforms covered, prompt panels, ChatGPT Ads monitoring, analytics, page audits and the MCP server. - [Pricing](https://www.moltmentions.com/pricing): Pay-as-you-go Wallet credits from $5, or Monitor from $6.03 a month for 25 prompts; credit, storage and attribution rates. - [Features](https://www.moltmentions.com/features): Prompt panels, AI visibility metrics and analytics, ChatGPT Ads monitoring, page audits and the MCP server, each with sample output. - [Use cases](https://www.moltmentions.com/use-cases): Two journeys step by step: your own data pipeline on the Wallet, or pre-built tracking on Monitor. - [About](https://www.moltmentions.com/about): Why MoltMentions exists, the founder's statement and who inspired it. ## Docs - [API reference](https://docs.thrivestack.ai/api-reference.html): Endpoints, authentication and request formats. - [MCP server](https://docs.thrivestack.ai/public/mcp/introduction.html): Connect an AI agent to the API. ## Optional - [Sign up](https://app.thrivestack.ai/auth/citedby/sign-up): Create an account and buy credits. - [Book onboarding](https://cal.com/thrivestack/citedby-ai-visibility): A free call to set up your first prompts. - [ThriveStack research](https://www.thrivestack.ai/research): Studies and guides on AI search visibility. ## Home page Page: [AI search API overview](https://www.moltmentions.com/) ### What this page covers - AI visibility pricing starts with $5 of pay-as-you-go credits, or Monitor from $6.03 a month. - Teams build six kinds of products on it, from white-label AI visibility reports to AI agents. - AI visibility tracking in a chat: ask your AI assistant, and MoltMentions does the work. - Most AI search APIs stop at the raw answer: MoltMentions is the only AI visibility API built for AI agents, from stored answers to CMO metrics and revenue. - A $5 AI visibility tool: pay for the answers you pull, with no monthly fee. - AI search monitoring in four API calls: prompts, answers, metrics and trends. - An AI search API tracks the answer, and a SERP API tracks the ranking. - Answer engine optimization guides for teams that build on AI search data. - You start with the AI search API in five steps, and the first takes minutes. - AEO, GEO and the AI search API, explained. - Start with the AI search API from MoltMentions. ### FAQ #### What does the MoltMentions page audit API check? It checks if AI bots can crawl a page, if the answer comes first, headings, schema markup, facts and sources, freshness and page speed. Add a prompt and it also tests if AI platforms cite the page. It returns a score out of 100 and a list of fixes. A page audit costs $0.23, so $5 audits 22 pages. #### What is an AI search API and what should it return? An AI search API is a programming interface that returns what AI platforms say about a brand. It should return the answer text, the brands named, the sources cited and a date, so you can track change week on week. #### Which AI platforms does the MoltMentions API cover? MoltMentions covers ChatGPT, Gemini, Claude, Perplexity and Google AI Overview, built on DataForSEO AI Optimization data. ChatGPT Ads monitoring, Google AI Mode and Copilot are live too. #### What can the MoltMentions prompt panel API tag and filter by? Each prompt carries tags for topic, keyword, persona, brand kit, language, country, buyer intent, and type (branded, non-branded or blended). You can run a panel for any tag or any mix of tags, on any platform. #### How quickly does the MoltMentions API return an answer? MoltMentions runs on DataForSEO. Live mode returns in up to 90 seconds for ChatGPT and Gemini, and the standard queue returns in up to 45 minutes at lower cost. #### How long does the MoltMentions API keep stored answers? You choose 1, 3, 6 or 12 months of retention. Stored answers are queryable by brand, prompt, platform, source and date through the API and the MCP server, and in the dashboard on Monitor. #### Can an AI agent use the MoltMentions API? Yes. The hosted MCP server has 54 tools, and 53 of them only read data. Connect it to Claude, Claude Code, Cursor or any MCP client. You sign in with OAuth or use a personal access token. #### What does ChatGPT Ads monitoring return? ChatGPT Ads monitoring shows which advertisers appear in ChatGPT answers for your prompts, with the ad text and the date. It is live now. It costs 25 credits a request, which is 5 times a ChatGPT search. #### How is answer engine optimization different from SEO? SEO works to rank a page in a list of links. Answer engine optimization works to get a brand named and cited inside the answer an AI assistant writes. The two use different data, so they need different APIs. #### Can the MoltMentions API show traffic and revenue from AI search? Yes. Attribution links visits from AI sources to leads and revenue, and it can be used on its own or with visibility data. #### Why should a brand monitor AI search every week? AI answers change from week to week. Google Ads data via DataForSEO, pulled Nov 6, 2026, shows monthly US searches for "ai visibility tool" rose from 590 in Nov 2025 to 1,300 in Aug 2026, so the topic is moving fast and one snapshot goes stale. #### How do I get started with the MoltMentions API? Create a MoltMentions account, create an API key in Settings, and call the first endpoint. You can also book free onboarding. Updated 2026-11-11. Endpoint names and launch pricing on this page are proposed and may change. ## Pricing page Page: [Pricing](https://www.moltmentions.com/pricing) > MoltMentions AI search API pricing starts with $5 of pay-as-you-go credits, or Monitor from $6.03 a month for 25 prompts. Credits, storage, attribution. Start with $5 of pay-as-you-go credits (Wallet), or Monitor from $6.03 a month (25 prompts). - Wallet (pay as you go): $5 minimum. Buys 12,285 credits, about $0.41 per 1,000 credits. - Monitor (monthly, by prompt volume): $6.03 a month, for 25 prompts. Use a beautiful UI to create, schedule, run and store prompts, review trends and recommendations, and understand revenue attribution added by events. - Attribution analytics: $20 per 100,000 events a month Launch pricing is proposed and may change. ### What would it cost you a month? | Runs | | | | Stored answers | | | | Visibility metrics | | | | Recommendations | | | | Attribution | | | | Retrievals | | | | Content page audits | | | ### Credits per request | Request | Credits | Per $5 | | Create a prompt by tags | 5 | 2,457 | | ChatGPT search | 5 | 2,457 | | ChatGPT full response (fan-out, ads, shopping) | 7 | 1,755 | | ChatGPT Ads monitoring (5x ChatGPT search) | 25 | 491 | | Owned and earned media recommendations, per stored answer | 10 | 1,228 | | Google AI Overview | 5 | 2,457 | | Google AI Mode | 4 | 3,071 | | Gemini | 4 | 3,071 | | Google Search | 3 | 4,095 | | Audit a page for AI readiness | $0.23 a page | 22 | | Content and technical recommendations (AEO, SEO, crawlers, entity schema) | Free with a page audit | Included | | Store a prompt | Free | No charge | | Store an answer (first month; each extra month adds 10%) | $0.001 an answer | 5,000 | | Retrieve a stored answer | $0.00005 an answer | 100,000 | ### Run and store, priced by prompt volume | Prompts tracked | Runs a month | Price a month | | 25 | 750 | $6.03 | | 250 | 7,500 | $60.26 | | 1,000 | 30,000 | $241.05 | | 5,000 | 150,000 | $1,205.25 | ### $20 per 100,000 events a month | Events a month | Price a month | | 100,000 | $20 | | 250,000 | $60 | | 1,000,000 | $200 | ### Store every answer in a format you can query, with the UI and the MCP server. | Retention | Per stored answer | Added to first month | 3,000 answers | | 1 month | $0.001 | base | $3.00 | | 3 months | $0.0012 | +20% | $3.60 | | 6 months | $0.0015 | +50% | $4.50 | | 12 months | $0.0021 | +110% | $6.30 | ### Pricing FAQ #### What recommendations does the MoltMentions Monitor plan give? Monitor gives recommendations for AEO, SEO, crawlers, entity schema, owned media and earned media. They are built from your stored AI answers and the pages you have audited. Content and technical recommendations (AEO, SEO, crawlers and entity schema) are free once a page is audited. Owned and earned media recommendations come from the citation gaps in your stored answers and cost 10 credits per stored answer. Launch pricing may change. #### What is the Monitor plan in the MoltMentions API? Monitor is a monthly plan for teams that track prompts on a schedule. You create prompts, schedule runs, run them and store every answer, and you pay by prompt volume. For example, 25 prompts run daily on ChatGPT search cost $6.03 a month, and 1,000 prompts cost $241.05. Attribution analytics is added by events. Plan names and prices are proposed and may change. #### How is attribution analytics priced in the MoltMentions API? Attribution analytics costs $20 per 100,000 events a month, billed in blocks of 100,000. Events combine website traffic (GA4), AI crawler server logs, billing events (Stripe) and CRM records that are matched and synced. For example, 250,000 events cost $60 a month. Launch pricing may change. #### How much does it cost to create and store prompts with the MoltMentions API? Creating a prompt from tags costs 5 credits, so $5 of credits (12,285) creates 2,457 prompts. Storing prompts is free. Running a stored panel costs the normal credits per request. For example, creating 25 prompts and running them once on ChatGPT search costs 125 plus 125 credits, which is about $0.10. Launch pricing may change. #### Do MoltMentions API credits expire? No. Credits never expire, so you can buy them when it suits you and use them whenever you need them. #### How does credit pricing work for the MoltMentions API? One credit balance is shared across all platforms. The Wallet plan is pay as you go. You buy credits, with a $5 minimum that gives 12,285 credits, and each request costs a set number of credits by platform. A ChatGPT search request costs 5 credits, so $5 covers 2,457 of them. There is no monthly fee. Launch pricing may change. #### How much does it cost to store AI answers for analytics? Storing an answer for the first month costs $0.001. Each extra month of retention adds 10% of that price, so 3 months is $0.0012, 6 months is $0.0015 and 12 months is $0.0021 per stored answer. The API and the MCP server are included (limits apply), and Monitor adds the dashboard. Launch pricing may change. ## About page Page: [About](https://www.moltmentions.com/about) > Why we built MoltMentions: AI search is exploding, teams ask Claude and ChatGPT instead of dashboards, and AI visibility tools cost too much. ### Why we're building this (Gururaj Pandurangi, founder) "It's time for the AI search industry to molt. Every brand deserves to know why AI isn't citing them, at rock-bottom prices, with no strings attached." - AI search is exploding. Buyers now ask ChatGPT, Gemini and Perplexity before they ever visit a website. Every brand and every agency has to cope with that, starting now. - Teams want answers, not another dashboard. People are leaving rigid SaaS screens behind. They ask Claude or ChatGPT a question and get informed in seconds, so the data has to live where they already work. - AI visibility has been priced out of reach. Profound, Peec, Searchable and 300 others put the same basic question behind tiered plans. Knowing what AI says about your brand shouldn't need a subscription, worse yet an enterprise contract. - It's time to shed the old leaves. That's the molt in MoltMentions. We want every brand to find out why it isn't visible, at rock-bottom prices, with no strings attached. ### Inspired by - Adam Robinson (in memory): Founder of RB2B and Retention.com. He showed me you can bootstrap, keep it simple and build something people find useful. https://www.linkedin.com/in/retentionadam/ Technical SEO: - Mike King: Founder of iPullRank. Treats search as relevance engineering, down to how machines retrieve and rank passages. https://www.linkedin.com/in/michaelkingphilly/ - Patrick Stox: Made crawling, indexing and canonicalization understandable, backed by studies of millions of sites. https://www.linkedin.com/in/patrickstox/ - Ross Hudgens: Founder of Siege Media. Shows how content and links compound into durable organic growth. https://www.linkedin.com/in/rosshudgens/ - Aleyda Solís: Founder of Orainti. Crawling Mondays and SEOFOMO keep the whole industry current. https://www.linkedin.com/in/aleyda/ - Lily Ray: The clearest voice on E-E-A-T, quality, and what algorithm updates actually reward. https://www.linkedin.com/in/lily-ray-44755615/ - Glenn Gabe: G-Squared Interactive. The go-to analyst of Google's core updates and their winners and losers. https://www.linkedin.com/in/glenngabe/ - Marie Haynes: Turned Google's quality guidelines into practical playbooks, and now applies them to AI search. https://www.linkedin.com/in/marie-haynes - Cyrus Shepard: Founder of Zyppy. Careful ranking studies that cut through SEO myth. https://www.linkedin.com/in/cyrusshepard/ AI search (AEO and GEO): - Ethan Smith: CEO of Graphite. A leading voice on answer engine optimization and on measuring what AI answers say. https://www.linkedin.com/in/ethanls/ - Rand Fishkin: Co-founder of SparkToro. Research on zero-click search and AI answers that shows where buyers really find information. https://www.linkedin.com/in/randfishkin/ - Dan Petrovic: DEJAN. Probes how LLMs perceive brands, with research tools like AI Rank. https://www.linkedin.com/in/seoguy/ - Britney Muller: Teaches marketers how LLMs actually work, and how to put AI to work in practice. https://www.linkedin.com/in/britneymuller/ - Chris Long: Co-founder of Nectiv. Practical AEO and AI search tactics, shared in public almost every day. https://www.linkedin.com/in/chris-long-marketing/ - Crystal Carter: Explains how LLMs and AI search find, read and describe brands, in plain language. https://www.linkedin.com/in/crystal-carter-digital/ - Mordy Oberstein: Makes the case that in AI search, brand is the strategy: answers favour brands people already know. https://www.linkedin.com/in/mordyoberstein/ None of these people is affiliated with MoltMentions. ### About FAQ #### Why is it called MoltMentions? To molt is to shed old skin. MoltMentions sheds rigid dashboards and tiered pricing so any brand can track its mentions in AI search, using Claude, ChatGPT or any other LLM. #### Who builds MoltMentions? MoltMentions is built by ThriveStack, founded by Gururaj Pandurangi. The hosted Monitor plan runs on MoltMentions. #### How much does MoltMentions cost? You can start with $5 of pay-as-you-go credits and no monthly fee, or use the Monitor plan from $6.03 a month for 25 prompts. Launch pricing is proposed and may change. #### Are the SEO experts on this page affiliated with MoltMentions? No. They are people whose public work taught and inspired the team. None of them is affiliated with, or endorses, MoltMentions. ## Use cases page Page: [Use cases](https://www.moltmentions.com/use-cases) > Two step-by-step journeys for the MoltMentions AI search API: run tagged prompts into S3 or BigQuery, or use built-in analytics with Slack and MCP. ### Scenario 1: Build an AI visibility product, or send a client weekly data Plan: Wallet (pay as you go). 1. Create a brand kit. The brand kit describes the brand you track. Every prompt and run refers back to it, so set it up first. 2. Build a set of prompts. Use the Prompt creation library to generate the questions your buyers ask. Tag each prompt and store it, because tags decide what runs together later. (Generate from the Prompt creation library; Tag by topic, persona, intent and market; Store the set) 3. Test a few prompts in Live mode. Run a handful of prompts live and read the real answers before you schedule anything. 4. Review the output with a skill. Open the results in Claude or ChatGPT and let a skill review them, so you see which prompts return useful answers and which do not. 5. Iterate until the prompts are right. Use what the review found to tune the set. Repeat steps 3 to 5 until the answers match what your buyers ask. (Delete prompts; Update prompts; Create prompts) 6. Create a destination for the output. Point the output at storage you own. (S3 bucket; BigQuery) 7. Schedule runs by tag in Claude. Create a scheduled task in Claude that runs specific prompts by tag on a regular basis. Pass the destination with every run. (Weekly prompts: standard mode, results land 10 min to 1 hr later, destination passed; Daily prompts: destination passed as a parameter) 8. Check the output. Open the S3 bucket or BigQuery table and confirm each scheduled run landed with its tags. 9. Show the analytics in your app, Claude Artifacts, slide decks, billboards… anywhere. Run a skill that reads the stored answers and puts the analytics wherever they are needed: your app, a Claude Artifact, a slide deck, even a billboard. Keep it current for your product or the client's weekly report. ### Scenario 2: Use the pre-built AI Visibility Analytics and Recommendations Plan: Monitor (monthly, by prompt volume). Steps 1 to 5 are the same as Scenario 1. 6. Use MoltMentions as the destination. Skip the bucket. Runs are stored in MoltMentions, where the pre-built AI Visibility Analytics and Recommendations read them. 7. Create a schedule in MoltMentions. Set which prompts run and how often inside MoltMentions. No scheduled task to maintain. 8. Connect Slack for notifications. Send notifications to a Slack channel, so the team hears about changes without opening a dashboard. 9. Manage it all through MCP. Use the MoltMentions MCP server to manage visibility, reporting and recommendations from your AI assistant. Illustrative sample. ### Use cases FAQ #### Which plan does each AI search API use case use? Scenario 1, your own pipeline, uses the Wallet plan: pay as you go, with credits you buy when you need them. Scenario 2, pre-built in MoltMentions, uses the Monitor plan: a monthly plan priced by prompt volume. See the pricing page for current launch pricing, which may change. #### Which AI search API use case should I start with? Start with Scenario 2 if you want reporting quickly: answers stay in MoltMentions, which runs the schedule, sends Slack notifications and gives analytics and recommendations. Pick Scenario 1 if you are building an AI visibility product or an agency report and need the answers in your own S3 bucket or BigQuery. #### Are the first five steps the same in both journeys? Yes. Both start by creating a brand kit, building and tagging prompts with the Prompt creation library, testing a few prompts in Live mode, reviewing the output with a skill in Claude or ChatGPT, and then deleting, updating or creating prompts until they are right. #### What is standard mode in the AI search API? Standard mode runs prompts with a delay: results arrive 10 minutes to 1 hour after the run. It suits prompts you track weekly, where an hour does not matter. #### How do I send AI search API results to my own warehouse? Create a destination for an S3 bucket or BigQuery, then pass it with each run, for example run_prompts_by_tag(tags, destination) from a scheduled task in Claude. Tool and endpoint names are illustrative. #### Can I manage MoltMentions from an AI assistant? Yes. Connect the MoltMentions MCP server to Claude, Claude Code or Cursor and manage visibility, reporting and recommendations from the assistant. ## Features page Page: [Features](https://www.moltmentions.com/features) > AI answers from every engine, prompt panels, stored answers, AI visibility metrics, ChatGPT Ads, page audits, recommendations, attribution and MCP. ### AI answers Track AI visibility across ChatGPT, Google AI Mode, AI Overviews, Gemini, Claude and Perplexity. One API. One response shape. Brands named, sources cited, date. ### Prompt panels Prompt tracking with prompt panels you create, store and run by tag. Create prompts from tags, store them once, and run only the slice you need. ### Stored answers Store every AI answer once: query it by brand, prompt, platform, source or date. Your own record of what AI said, week after week. Keep answers 1, 3, 6 or 12 months; retrieving one costs $0.00005. ### AI visibility metrics Stored answers automatically calculate your share of voice and citations in real time. One snapshot shows where you are. A weekly record shows where you are heading. Share of voice, rank, mentions and citations on every stored answer: $0.005 an answer, included in Monitor. ### ChatGPT Ads monitoring ChatGPT Ads monitoring shows which advertisers appear on your prompts. See which rival pays to appear on your prompts, with what ad, on which day. Live now. ### Page audit One page audit call scores a page’s AI readiness. Send a URL and get back a score out of 100, the fixes to make, and whether AI engines cite the page today. Live now. ### Recommendations AEO and GEO recommendations: what to fix on your pages, and where to get cited. Each recommendation reads what AI said about you and what your pages show, and returns a ranked list of fixes. ### Attribution tracking AI traffic attribution: tie AI visits, sign-ups and revenue to the answers that sent them. See whether AI search makes you money, in three steps. ### MCP server The MCP server puts AI visibility data inside your AI agent. 54 tools, 53 read only. Sign in with OAuth or a token. Setup docs ### Features FAQ #### What features does MoltMentions have? AI answers from ChatGPT, Google AI Mode, AI Overviews, Gemini and more in one response shape; prompt panels you create from tags; stored answers you can query; AI visibility metrics such as share of voice and citations; ChatGPT Ads monitoring; page audits for AI readiness; content, technical, owned and earned media recommendations; attribution that ties AI visits to revenue; and an MCP server so Claude, ChatGPT or any MCP client can use all of it. #### Which MoltMentions features are live today? All of them: AI answers from every engine, prompt panels, stored answers, AI visibility metrics, ChatGPT Ads monitoring, page audits, recommendations, attribution tracking and the MCP server. #### Do I need the Monitor plan to use these features? No. On the Wallet you pay per request from credits and call each feature through the API or the MCP server. Monitor adds the schedule, the app and AI visibility metrics on every stored answer, priced by prompt volume. #### Can I use MoltMentions features from Claude or ChatGPT? Yes. The hosted MCP server has 54 tools, and 53 of them only read data. Connect it to Claude, Claude Code, Cursor or any MCP client and ask for runs, metrics, audits or recommendations in plain language.