# ListeningMind

> Search-intent data for AI agents: what people in the US, Japan and Korea search for, in what order, and why. Call it over MCP, REST or WebMCP; people can run the same tools on this site.

News: the free whats_new tool (MCP) and the feed at /feed.xml.

## Start here
- One question, one result: `POST https://daas.listeningmind.ai/v1/outcomes/plan` with `{"question": "..."}` (any of ko, en, ja; free). It reuses a stored analysis when one fits, shows the price, time and a free preview; `POST /v1/outcomes/runs` with `plan_id`, `approve` and `max_lima` runs it through the order engine; `GET /v1/outcomes/runs/{run_id}/result` returns the brief (JSON, or `?format=md`) with evidence types OBSERVED, DERIVED, INFERRED and RECOMMENDATION, the quality status, limits and sources. Same over MCP (ask, then place_order with plan_id, then get_order_result with run_id), A2A and on the web at https://daas.listeningmind.ai/en/ask
- [MCP server](https://daas.listeningmind.ai/mcp): Streamable HTTP (protocol 2025-03-26 to 2026-07-28). Free tools: catalog, whats_new, estimate_cost, get_sample, topic_search; request_key for a key; the data tools below. Cline: add `"type": "streamableHttp"` next to the url, or it tries SSE
- [OpenAPI](https://daas.listeningmind.ai/openapi.json): REST contract. Send keys as `Authorization: Bearer <key>` or `LM-API-KEY: <key>`
- [Agent card](https://daas.listeningmind.ai/.well-known/agent-card.json): A2A concierge: plans a market question on the outcome engine and runs it when approved
- [API catalog](https://daas.listeningmind.ai/.well-known/api-catalog) and [MCP server card](https://daas.listeningmind.ai/.well-known/mcp/server-card.json) · public MCP repository: https://github.com/listeningmind/listeningmind-mcp
- Every page answers `Accept: text/markdown` with a plain version, e.g. [https://daas.listeningmind.ai/en/tools/keyword_info](https://daas.listeningmind.ai/en/tools/keyword_info)

## Order data: quote, order, delivery
- Three tiers: core = the 15 intent elements of a topic (`intent_kg`); deep = CEP map (`cep_detail`), taste (`taste`), traffic leak paths (`funnel_leak`); raw = raw search tools
- 15 elements = decision layer (CEP · Buying situation, Occasion, Nano intent, KBF · Choice criteria, Constraint, Threshold, Avoidance, RTB · Reason to believe) + knowledge layer (Heuristic, Convention, Expectation, Pitfall, Trade-off, Benchmark, Stance); the entity layer comes with them
- Flow: `GET https://daas.listeningmind.ai/v1/products` → `GET /v1/library?product=&market=&q=` (stored samples, free preview) → `POST /v1/quotes` → `POST /v1/orders` with `quote_id` and `max_lima` → `GET /v1/orders/{id}/wait?timeout=50` until delivered → `GET /v1/orders/{id}/result`
- MCP tools (15): ask, catalog, topic_search, get_sample, estimate_cost, request_key, my_account, keyword_info, intent_finder, path_finder, cluster_finder, ai_overview, intent_elements, place_order, get_order_result. Older names (list_products, find_samples, get_quote, whats_new, topic_build, intent_graph, run_outcome, outcome_status, order_status, wait_for_order, cancel_order, my_balance, my_notices, list_my_orders, buy_lima) still work
- A quote fixes price (Lima), stored or new build, as-of date and age, ETA, quality rule, ownership and budget left. Show its `say_to_user` to the person
- Prices: topic 70 Lima. The first 2 stored results per approving email are free. Stored results are sold up to 90 days old at the same price as a new build; the quote also offers rebuilding at the same price. Older ones are rebuilt when ordered. Reading something you bought is free
- A build that misses the quality rule is delivered and not charged. Topic builds take 2 to 6 minutes
- Ownership: category orders become shared samples; brand orders are private to the ordering key
- Delivery: poll, long-poll (`/wait`, up to 50 s), signed webhook (`callback_url`, header `LM-Signature: t=..,v1=hmac`), status page `/o/{id}`, notices (`GET /v1/notifications`) including newer versions of results you bought. A2A: `tasks/get` with the order code as the task id
- Keys work by header alone, with or without an MCP session: `Authorization: Bearer <key>` or `LM-API-KEY: <key>`. Clients on the stateless 2026-07-28 protocol have no session, so set the header in the client's MCP config
- In a chat (Claude, ChatGPT) on the session protocol (2025-11-25 and earlier): call request_key and show the link; after the person approves, the key is bound to this MCP session (Mcp-Session-Id), nothing to copy. Then estimate_cost → place_order → get_order_result (wait=50) in the same conversation. Give the person the status page link from say_to_user; in a later chat, pass that link (or the code plus its t= token) as order_id to get_order_result
- Stored topics: [Topic library](https://daas.listeningmind.ai/en/topics), `GET https://daas.listeningmind.ai/v1/topics?q=&market=` (free search), [llms-full.txt](https://daas.listeningmind.ai/llms-full.txt), [feed.xml](https://daas.listeningmind.ai/feed.xml)

## Data tools (raw data)
- [keyword_info](https://daas.listeningmind.ai/en/tools/keyword_info): Monthly volume trend, competition and CPC, and searcher gender and age (Korea). 13 Lima per 100 keywords (the data a call really returns; typical call about 1 Lima)
- [intent_finder](https://daas.listeningmind.ai/en/tools/intent_finder): Related searches for a keyword, ordered by volume. 1 Lima per call + 26 Lima per 1,000 results (the data a call really returns; typical call about 1 Lima)
- [path_finder](https://daas.listeningmind.ai/en/tools/path_finder): The searches people actually made before and after this one. 1 Lima per call + 520 Lima per 1,000 keywords in the paths (the data a call really returns; typical call about 95 Lima)
- [cluster_finder](https://daas.listeningmind.ai/en/tools/cluster_finder): Searches grouped into intent clusters: how consumers frame the category. 2 Lima per call + 650 Lima per 1,000 keywords in the clusters (the data a call really returns; typical call about 182 Lima)
- [ai_overview](https://daas.listeningmind.ai/en/tools/ai_overview): Top results, the AI overview text and the sources it cites. 2 Lima per keyword (the data a call really returns; typical call about 2 Lima)
- [trend_finder](https://daas.listeningmind.ai/en/tools/trend_finder): Searches whose volume grew fastest within a period. 1 Lima per call + 26 Lima per 1,000 issues (the data a call really returns; typical call about 4 Lima) (planned, not callable yet)

## Free, no key
- Always free (5): catalog, whats_new, estimate_cost, get_sample, topic_search. request_key is free too: it is how a person approves a key
- [catalog](https://daas.listeningmind.ai/v1/catalog): every tool with price, markets and input schema
- estimate_cost: `POST https://daas.listeningmind.ai/v1/estimate_cost` with `{"tool": ..., "arguments": {...}}`
- [samples](https://daas.listeningmind.ai/v1/samples): real saved responses from Intent KG runs, by tool and market

## Unit: Lima
- Prices are in Lima (리마, リマ), KRW 100. Raw tools are charged for the data a call really returns, at Data API rates (1 credit = KRW 1.3), rounded up to a whole Lima per call; ai_overview is 2 Lima per keyword. Call estimate_cost first and send max_lima: the charge never goes above it. listeningmind.com plan Lima cannot pay for these tools
- Responses report `cost_detail.total_cost` in Lima, with `unit: "lima"`

## Free allowance
- Every account (the e-mail that approves a key): 10,000 credits (130 Lima) a month, refilled on the 1st of each month (Korea time), not carried over
- A company e-mail account (not a free mail service): 100,000 credits (1,300 Lima) once with its first key; once per person; another address of the same person does not count
- Free Lima is used before paid Lima. The amounts are fixed in credits, so they stay the same if the credit price changes
- Free tools need no key and do not use the allowance: catalog, estimate_cost, get_sample, topic_search, whats_new
- `GET /v1/account` shows `free.monthly_left`, `free.signup_bonus_left` and `free.resets`

## Keys and budget
- `POST https://daas.listeningmind.ai/v1/keys/request` with `{"agent_name": ...}` returns an approval link and code for a person, and a poll URL
- The person approves once and sets a monthly budget in Lima. Calls above `max_lima` or the budget return 402 and are not run
- Errors are `application/problem+json` with a `next` field that says what to do

## Markets
- us: Google · jp: Google · kr: Naver + Google. Pass `market` on every call

## The ListeningMind apps (each one is its own MCP server, same account)
- Directory for people: https://daas.listeningmind.ai/en/apps · definitions: https://daas.listeningmind.ai/en/glossary. Every app has /mcp, /llms.txt, /agents.md, WebMCP on its pages (this hub too: https://daas.listeningmind.ai/agents.md), and the tools family_apps, recommend_app and handoff to find and continue in the others.
- From this site (WebMCP on every page, or the hub endpoints): site_guide, family_apps (https://daas.listeningmind.ai/_lm/family.json), recommend_app (https://daas.listeningmind.ai/_lm/recommend?goal=), app_guide, app_tools and call_app_tool (POST https://daas.listeningmind.ai/_lm/family/call with app, tool, args), confirmation_status, my_runs, glossary (https://daas.listeningmind.ai/en/glossary.json), read_this_page, open_app, go_to. Free reads answer at once; a run comes back as needs_confirmation with a confirm_url the signed-in member approves.
- [CEP Finder](https://daas-cep.listeningmind.ai/llms.txt): In which situations do people start looking for this category (category entry point map) (MCP `https://daas-cep.listeningmind.ai/mcp`)
- [Taste Finder](https://daas-taste.listeningmind.ai/llms.txt): In what sensory and emotional words do people choose this (consumer taste report from search data) (MCP `https://daas-taste.listeningmind.ai/mcp`)
- [Brand Finder](https://daas-brand.listeningmind.ai/llms.txt): Which brands lead search by generation or field, and what their brand assets are (MCP `https://daas-brand.listeningmind.ai/mcp`)
- [Pulse Finder](https://daas-pulse.listeningmind.ai/llms.txt): What is rising in search for a gender and age segment right now (MCP `https://daas-pulse.listeningmind.ai/mcp`)
- [Funnel Leak Finder](https://daas-funnel.listeningmind.ai/llms.txt): Where category search journeys leak to competitors (MCP `https://daas-funnel.listeningmind.ai/mcp`)
- [Demand Forecast](https://daas-forecast.listeningmind.ai/llms.txt): What category demand looks like in the next 3-6 months and when to launch or campaign (MCP `https://daas-forecast.listeningmind.ai/mcp`)
- [Creative Finder](https://daas-creative.listeningmind.ai/llms.txt): Finds purchase situations (CEP) in search data and drafts a creative, an image and a video for each (MCP `https://daas-creative.listeningmind.ai/mcp`)
- [Brand Naming Agent](https://daas-naming.listeningmind.ai/llms.txt): Creates and checks brand name candidates grounded in search data (MCP `https://daas-naming.listeningmind.ai/mcp`)
- [AOO](https://daas-aoo.listeningmind.ai/llms.txt): Measures whether AI agents can get real tasks done on a website (MCP `https://daas-aoo.listeningmind.ai/mcp`)
- [GEO Audit](https://daas-geo.listeningmind.ai/llms.txt): Checks whether a web page is ready to be read and cited by search engines and AI answer engines (MCP `https://daas-geo.listeningmind.ai/mcp`)
- [Prompt Builder](https://daas-prompt.listeningmind.ai/llms.txt): Finds what consumers ask AI and builds managed brand prompts with AI visibility (MCP `https://daas-prompt.listeningmind.ai/mcp`)
- [ACO](https://daas-aco.listeningmind.ai/llms.txt): Diagnoses whether a product detail page can be cited by AI answers and used by shopping agents (MCP `https://daas-aco.listeningmind.ai/mcp`)
- [Brand Wiki](https://daas-wiki.listeningmind.ai/llms.txt): Turns a brand website into the entities, wiki pages and evidence AI assistants read (MCP `https://daas-wiki.listeningmind.ai/mcp`)
- [Smart Store Suite](https://daas-store.listeningmind.ai/llms.txt): Which Smart Store products to grow, and how to rewrite their titles and detail pages in searchers' words (MCP `https://daas-store.listeningmind.ai/mcp`)
- [ICSS](https://daas-icss.listeningmind.ai/llms.txt): Synthetic consumer panel conditioned on search intent, for surveys and FGDs (MCP `https://daas-icss.listeningmind.ai/mcp`)
- [Intent KG](https://daas-intent.listeningmind.ai/llms.txt): What people look for about a topic, why, and how they choose (Intent Knowledge Graph) (MCP `https://daas-intent.listeningmind.ai/mcp`)

## Catalog: apps, agents, skills, MCP servers and APIs
- [Catalog](https://daas.listeningmind.ai/en/catalog) (also https://daas.listeningmind.ai/ko/catalog and https://daas.listeningmind.ai/ja/catalog; every catalog page answers `?format=md`). Outside skills, MCP servers and APIs are re-picked every Friday 19:00 KST from Smithery, the official MCP Registry, GitHub, SkillsMP and APIs.guru, scored on popularity, trust, search demand and search and AI Overview presence: [how we pick](https://daas.listeningmind.ai/en/catalog/method). Last update: 2026-10-23
- [MCP servers](https://daas.listeningmind.ai/en/catalog/mcps), top 10 this week: [Brave Search](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-brave), [GitHub](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-github), [Paper Search](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-adamamer20-paper-search-mcp-openai), [PubMed](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-pubmed), [Exa Search](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-exa), [Context7](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-upstash-context7-mcp), [Gmail](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-gmail), [Slack](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-slack), [Google Calendar](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-googlecalendar), [Google Drive](https://daas.listeningmind.ai/en/catalog/mcps/x-mcp-googledrive)
- [Skills](https://daas.listeningmind.ai/en/catalog/skills), top 10 this week: [humanizer](https://daas.listeningmind.ai/en/catalog/skills/x-skill-blader-humanizer), [open-design](https://daas.listeningmind.ai/en/catalog/skills/x-skill-nexu-io-open-design), [open-code-review](https://daas.listeningmind.ai/en/catalog/skills/x-skill-alibaba-open-code-review), [SkillSpector](https://daas.listeningmind.ai/en/catalog/skills/x-skill-nvidia-skillspector), [scientific-agent-skills](https://daas.listeningmind.ai/en/catalog/skills/x-skill-k-dense-ai-scientific-agent-skills), [SkillOpt](https://daas.listeningmind.ai/en/catalog/skills/x-skill-microsoft-skillopt), [github](https://daas.listeningmind.ai/en/catalog/skills/x-skill-openclaw-openclaw-github), [cognee](https://daas.listeningmind.ai/en/catalog/skills/x-skill-topoteretes-cognee), [slack](https://daas.listeningmind.ai/en/catalog/skills/x-skill-openclaw-openclaw-slack), [grill-me](https://daas.listeningmind.ai/en/catalog/skills/x-skill-mattpocock-skills-grill-me)
- [APIs](https://daas.listeningmind.ai/en/catalog/apis), top 10 this week: [Poly API](https://daas.listeningmind.ai/en/catalog/apis/x-api-googleapis-com-poly), [Travel Partner API](https://daas.listeningmind.ai/en/catalog/apis/x-api-google-com), [Deal API](https://daas.listeningmind.ai/en/catalog/apis/x-api-ebay-com-buy-deal), [Slack Web API](https://daas.listeningmind.ai/en/catalog/apis/x-api-slack-com), [They Said So Quotes API](https://daas.listeningmind.ai/en/catalog/apis/x-api-quotes-rest), [Stripe API](https://daas.listeningmind.ai/en/catalog/apis/x-api-stripe-com), [WhatsApp Business API](https://daas.listeningmind.ai/en/catalog/apis/x-api-whatsapp-local), [The Plaid API](https://daas.listeningmind.ai/en/catalog/apis/x-api-plaid-com), [Braket](https://daas.listeningmind.ai/en/catalog/apis/x-api-amazonaws-com-braket), [DocuSign REST API](https://daas.listeningmind.ai/en/catalog/apis/x-api-docusign-net)
- [Apps](https://daas.listeningmind.ai/en/catalog/apps) · [Agents](https://daas.listeningmind.ai/en/catalog/agents) · [Register your skill, MCP server or API](https://daas.listeningmind.ai/en/catalog/submit)

## Wiki
- [ListeningMind wiki](https://daas.listeningmind.ai/en/wiki) (also /ko/wiki and /ja/wiki; every page answers `?format=md`): 71 pages in English
- [What is AX (AI Transformation)?](https://daas.listeningmind.ai/en/wiki/ax): AX (AI Transformation) is the process of identifying parts of a company's work that AI Agents can handle, and providing them with the data to read, tools to call, procedures to follow, and budgets they can use. Humans handle judgment and approvals.
- [Enterprise Marketing AX: Starting AI Transformation with Search Data](https://daas.listeningmind.ai/en/wiki/marketing-ax): Enterprise Marketing AX transforms marketing tasks such as market research, audience definition, messaging, content, advertising, and AI search readiness. AI agents perform the work using supporting data while humans make decisions. The supporting data comes from actual customer search records.
- [What Are Agent Skills: Passing Work Procedures to Agents as Files](https://daas.listeningmind.ai/en/wiki/skill): A skill is a package that bundles the procedures, criteria, and examples for an AI agent to follow when performing a specific task into a SKILL.md file and supporting files. While MCP tells an agent what it can do, a skill tells it in what order and by what criteria to do it.
- [Intent Intelligence: Turning Search Behavior into Shared Evidence](https://daas.listeningmind.ai/en/wiki/intent-intelligence): Intent intelligence is the data and methodology that organizes search behavior into a structure of what people look for, why they search, and how they choose, allowing humans and AI agents to make decisions based on the same evidence.
- [Search Data API: 6 Tools to Connect Search Intent to SYSTEM and Agents](https://daas.listeningmind.ai/en/wiki/api): The ListeningMind Data API is a REST API that returns search volume, search paths, related searches, keyword clusters, AI Overviews, and trending topics for South Korea, the United States, and Japan in JSON. The same tools are also available via MCP so AI agents can call them directly.
- [What Is MCP: Open Standard for AI Agents to Call Tools and Data](https://daas.listeningmind.ai/en/wiki/mcp): Model Context Protocol (MCP) is an open standard that lets an AI Agent discover and call external tools and data in a uniform way. When a server exposes its capabilities as tools, clients such as Claude or ChatGPT can call those tools and use the returned results.

## Optional
- [Buy](https://daas.listeningmind.ai/en/buy): MCP, Skills, DaaS API and orders compared, with prices. 1 Lima = KRW 100 at listeningmind.com Lima pack prices (VAT excluded). listeningmind.com plans: Basic KRW 720,000/month with 360 Lima; Standard KRW 2,400,000/month with 2,400 Lima; Advance KRW 7,200,000/month with 14,400 Lima. Free on .ai: 130 free Lima every month · 1,300 more once for a company e-mail (see Free allowance above)
- [Skills](https://daas.listeningmind.ai/en/skills): 4 downloadable skills that run on this MCP server (Lima). Separately, [lm-skills on GitHub](https://github.com/ListeningMind-Daas/lm-skills) (MIT): 12 task skills and a skill creator that call the ListeningMind DaaS Data API with a DaaS API key · [DaaS API](https://daas.listeningmind.ai/en/enterprise): integration levels, usage estimator, inquiry
- [Showcase](https://daas.listeningmind.ai/en/apps): 16 apps built on these tools and what they return
- Korean: https://daas.listeningmind.ai/ko/ · Japanese: https://daas.listeningmind.ai/ja/