Intentlink

io.github.whalefuntech由 GitHub 帳號 whalefuntech 發布v1.0.0更新於 Oct 27, 2026

The best agent monetization tool: free search, clear rev-share, 10M+ commerce/travel/local offers.

已驗證Streamable HTTP可網頁執行位置與生活商業與電商

概覽

AI 產生的概覽

IntentLink 是遠端 MCP 伺服器,讓助理搜尋聯盟購物、旅遊與活動優惠,並回傳可追蹤的購買或預訂連結。

可以做什麼
  • 以自然語言查詢搜尋購物商品,取得排序後的推薦,並可用關鍵字與商品類型篩選
  • 依城市搜尋飯店、機票、套裝行程、機場接送與租車等旅遊優惠,回傳可追蹤的預訂連結
  • 搜尋演唱會、體育賽事、劇場等現場票務活動,回傳排序後的活動組合
  • 讀取旅遊城市資源,取得旅遊搜尋可用的城市值
適用情境
適合需要把購物者或旅客的需求轉成具體商業推薦並附上可點擊連結的情境,也適合營運方希望把聯盟歸因綁定到自己的 agent ID 時使用。
執行需求
需要遠端 MCP 端點,不需本機套件或執行環境。需要從服務商入口網站取得 agent_id,並在每次工具呼叫時傳入,而非寫在靜態設定中。需要能連線至該託管端點的網路。
安裝前請注意
回傳結果是帶商業利益的聯盟推薦,建議可能受佣金影響;README 說明轉換收益中營運方可保留 80%。agent_id 於每次工具呼叫中傳遞,歸因會回傳至該 ID。README 稱不儲存個人身分資訊、僅以暫時性對話內容進行比對,但使用者查詢仍會傳送給第三方服務。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Intentlink,將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable,透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式,或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南,先向你展示發布者、端點和登入方式,經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.whalefuntech%2Fintentlink/install.md,並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前,先告訴我它的發布者、端點和登入方式,等我確認。不要把 API 金鑰寫進命令或我會提交的檔案;問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "intentlink": {
      "type": "http",
      "url": "https://api.intentlink.io/mcp/"
    }
  }
}

README

IntentLink

Monetize your AI agent's conversations — natively, without banners.

IntentLink is a commercial intent network built for the AI agent era. Add the IntentLink MCP server (or REST API) to your agent, and it gains the ability to surface relevant, monetizable commerce recommendations directly inside its answers — not as an ad unit bolted on top, but as part of the conversation itself.

Website · Docs · Developer Dashboard · For Merchants & Partners

Why IntentLink

  • Native to the conversation. Recommendations surface as part of the agent's answer — contextual and relevant, not a banner or pop-up.
  • Zero UX disruption. Your agent stays your agent. No iframes, no forced redirects, no interruption of the conversation flow.
  • You keep 80% of revenue. Transparent rev-share on every converted intent. No upfront fee, no platform fee — IntentLink only earns when you do.
  • Privacy-first by design. No PII is stored. Intent matching runs on transient conversation context only, aligned with GDPR/CCPA.
  • Works with any MCP-compatible runtime — Claude, OpenAI-compatible agents, LangChain, or custom stacks (a language-agnostic REST API is available as a fallback).

Quick Start

Add IntentLink as a remote MCP server:

json
{"mcpServers": {"intentlink": {"url": "https://api.intentlink.io/mcp/"}}}

Your agent_id is passed per tool call, not in the static config — get one at portal.intentlink.io.

Prefer REST? See the full docs for the language-agnostic API.

What it does

Once connected, your agent can match user intent to relevant commercial offers and return a trackable purchase link for the user to complete the transaction — attribution flows back to your agent_id automatically.

About this repo

This repository hosts public documentation and integration resources for IntentLink. The IntentLink MCP server itself is remotely hosted (not distributed as an installable package) — see Quick Start above to connect.

License

MIT — see LICENSE file for details.

來源:README.md,提交 c541e92

工具

3

search_products

Search affiliate shopping products and get ranked recommendations. Only `query` is required — a natural-language description of what the shopper wants (e.g. "a warm waterproof jacket for winter hiking"). Add fields to narrow results: - keywords: exact product terms. Use a SPECIFIC product type ("women running shoes", "stainless steel knife set"), not a bare generic noun ("shoes", "knife") — a generic keyword can surface the wrong audience. Audience/occasion go in `query`. - commerce_l2s: category ids to restrict to (read the `commerce://categories` resource for valid values — that taxonomy is for shopping only). - max_price / min_price: price band (USD). platforms: ["amazon"|"walmart"] (empty=all). - intent: ranking preset — "cheapest" | "best_discount" | "top_rated" | "best_value". - require_commission: only return products that pay a commission (default: server setting, normally on). Pass false to include zero/unknown-commission products. - limit: max results (default 10). `agent_id` is REQUIRED: your registered, active agent id. It is the attribution key and the access key — a missing/blank or unregistered agent_id is rejected (no anonymous use). Returns {count, products:[{offer_id,title,brand,price,rating,...,buy_url,product_url}]}. Each product's `buy_url` is ALREADY the trackable affiliate buy link — hand it to the shopper directly. (`product_url` is the plain product page.) There is no separate resolve/click step.

輸入參數limitquery必填intentagent_id必填keywordsmax_pricemin_priceplatformscommerce_l2srequire_commission

search_travel

Search travel offers and get trackable booking links for a given product type + city. REQUIRED: - type: which travel product — one of ["hotel","flights","bundle","airport_transfer","car_rental"]. - city: city name from the supported list — read the `travel://cities` resource for valid values (that city list is for travel only). Pass hotel.state (e.g. "NY") when the city name is ambiguous. - agent_id: your registered, active agent id — the attribution + access key. A missing/blank or unregistered agent_id is rejected (no anonymous use). HOTEL runs a live Trip.com search — pass the `hotel` object with the stay: check_in / check_out (YYYY-MM-DD), rooms, adults, children, state. If those are missing/invalid or the city can't be resolved, you get the hotel default links instead. Other types currently return curated default links (the `hotel` object is ignored for them). Returns {status, reason, message, links:{<type>:[{title, brand, platform, buy_url, product_url, ...}]}}. status is "ok" = a live hotel result; "fallback" = curated default links (see `reason`); "no_results" = nothing available for the type; "error" = the agent_id was rejected. Each `buy_url` is ALREADY the trackable booking link — hand it to the traveler directly. `product_url` is the plain merchant page.

輸入參數city必填type必填hotellimitagent_id必填

search_events

Search live ticketing/events (concerts, sports, theater) and get ranked event bundles. Only `query` is required — a natural-language description of what the buyer wants (e.g. "taylor swift concert near LA this summer"). Add fields to narrow results: - keywords: exact terms — performer / team / show / venue (a precise lookup). - categories: e.g. ["Concerts"|"Sports"|"Theater"] (read the `events://categories` resource for valid values). - city / state: location (e.g. city="Los Angeles", state="CA"). country defaults to US. - date_from / date_to: explicit date window, YYYY-MM-DD (event dates are date-only). - when: fuzzy time — "this_weekend" | "this_week" | "this_month" | "next_7_days" | "next_30_days" (ignored if date_from/date_to are given). - min_price / max_price: ticket 'from' price band (USD). - sort: "relevance" (default) | "soonest" | "cheapest". - limit: max results (default 10). `agent_id` is REQUIRED: your registered, active agent id — the attribution + access key. A missing/blank or unregistered agent_id is rejected (no anonymous use). Returns {count, message, events:[{bundle_id,title,venue,city,next_date,price_min,..., sessions:[{date,price,buy_url,landing_url}]}]}. Each session's `buy_url` is ALREADY the trackable ticket link — hand it to the buyer directly. `landing_url` is the plain event page. When a date filter is given, `sessions` are the in-window dates.

輸入參數citysortwhenlimitquery必填statedate_toagent_id必填keywordsdate_frommax_pricemin_price另有 1 個

版本歷史

1
  1. v1.0.0最新Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分