Apify actor를 통해 Google Search, Maps, 뉴스, 소셜 미디어, 이커머스 등 공개 웹 데이터에 AI agent가 접근할 수 있도록 지원합니다.
다른 이름 subimpact/data-mcp, data-mcp, @subimpact/data-mcp
- 스타
- 0
- Tools
- —
- 다운로드, 30일
- —
- 최신 버전
- 1.2.0
- 마지막 커밋
- 2026년 10월 29일
- 라이선스
- MIT
subimpact data MCP란?
Apify actor를 통해 Google Search, Maps, 뉴스, 소셜 미디어, 이커머스 등 공개 웹 데이터에 AI agent가 접근할 수 있도록 지원합니다.
프로젝트 README에서 생성된 요약
subimpact data MCP server 정보
웹에서 실시간 정보를 수집해야 하는 AI agent는 이 server를 사용하여 다양한 공개 데이터 소스에 접근할 수 있습니다. 여러 탭을 직접 브라우징하는 대신, agent는 Google Search, Google Maps, YouTube, TikTok은 물론 Substack이나 Bluesky와 같은 니치 플랫폼으로부터 구조화된 JSON 데이터를 가져올 수 있습니다. 이 server는 API token을 사용하여 사용자의 Apify 계정에 연결되므로, 모든 스크래핑 작업은 로컬이 아닌 Apify 인프라에서 실행됩니다.
개발자는 필요한 환경 변수를 설정하여 Claude Desktop, Claude Code 또는 Cursor와 통합할 수 있습니다. 이 server는 일반 검색 결과 및 AI Overviews를 위한 google_search, 동남아시아 퍼블리셔 추적을 위한 news_mentions, 소셜 프로필 및 이메일 추출을 위한 website_contacts와 같은 특화된 도구를 제공합니다. 예상치 못한 비용 발생을 방지하기 위해 모든 도구 호출에는 _maxChargeUsd 파라미터를 통해 0.50 USD의 기본 지출 한도가 적용됩니다. 또한 사용자는 _maxItems 설정을 사용하여 모델에 반환되는 데이터 양을 제한함으로써 컨텍스트 윈도우 제한을 관리할 수 있습니다.
활용 사례
- 1검색 결과를 통한 시장 조사AI agent가 Google 검색을 수행하여 일반 검색 결과, 지식 그래프 및 AI Overviews를 구조화된 형식으로 가져올 수 있습니다.
- 2웹사이트를 통한 리드 생성개발자가 agent에게 특정 도메인 목록에서 연락처 정보와 소셜 미디어 링크를 찾도록 지시할 수 있습니다.
- 3경쟁사 소셜 미디어 모니터링agent가 TikTok에서 특정 해시태그를 스크래핑하여 영상 참여도와 크리에이터 프로필을 분석할 수 있습니다.
- 4로컬 비즈니스 인텔리전스AI가 평점, 리뷰, 영업시간을 포함한 지역 업체에 대한 상세 정보를 수집할 수 있습니다.
프로젝트의 README를 바탕으로 작성되었습니다. 모든 활용 사례는 해당 내용의 인용문과 대조하여 확인합니다.
subimpact data MCP 설치 방법
형식 확인: 2026년 11월 10일Cursor에 추가Goose에 추가Claude Code에 subimpact data MCP 설치
$ claude mcp add --env 'APIFY_TOKEN=<YOUR_APIFY_TOKEN>' data-mcp -- npx -y @subimpact/data-mcpCursor에 subimpact data MCP 설치
~/.cursor/mcp.json
{
"mcpServers": {
"data-mcp": {
"command": "npx",
"args": [
"-y",
"@subimpact/data-mcp"
],
"env": {
"APIFY_TOKEN": "<YOUR_APIFY_TOKEN>"
}
}
}
}VS Code에 subimpact data MCP 설치
.vscode/mcp.json
{
"servers": {
"data-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@subimpact/data-mcp"
],
"env": {
"APIFY_TOKEN": "<YOUR_APIFY_TOKEN>"
}
}
}
}VS Code CLI에 subimpact data MCP 설치
$ code --add-mcp '{"name":"data-mcp","type":"stdio","command":"npx","args":["-y","@subimpact/data-mcp"],"env":{"APIFY_TOKEN":"<YOUR_APIFY_TOKEN>"}}'Claude Desktop에 subimpact data MCP 설치
claude_desktop_config.json
{
"mcpServers": {
"data-mcp": {
"command": "npx",
"args": [
"-y",
"@subimpact/data-mcp"
],
"env": {
"APIFY_TOKEN": "<YOUR_APIFY_TOKEN>"
}
}
}
}Codex에 subimpact data MCP 설치
$ codex mcp add data-mcp --env 'APIFY_TOKEN=<YOUR_APIFY_TOKEN>' -- npx -y @subimpact/data-mcpCodex config.toml에 subimpact data MCP 설치
~/.codex/config.toml
[mcp_servers.data-mcp] command = "npx" args = ["-y", "@subimpact/data-mcp"] [mcp_servers.data-mcp.env] APIFY_TOKEN = "<YOUR_APIFY_TOKEN>"
Gemini CLI에 subimpact data MCP 설치
$ gemini mcp add -s user -e 'APIFY_TOKEN=<YOUR_APIFY_TOKEN>' data-mcp npx -- -y @subimpact/data-mcpGemini settings.json에 subimpact data MCP 설치
~/.gemini/settings.json
{
"mcpServers": {
"data-mcp": {
"command": "npx",
"args": [
"-y",
"@subimpact/data-mcp"
],
"env": {
"APIFY_TOKEN": "<YOUR_APIFY_TOKEN>"
}
}
}
}구성
APIFY_TOKEN필수비밀Your Apify API token. Runs are billed to your Apify account at each actor's Store price.
| 이름 | 설명 | 유형 |
|---|---|---|
| APIFY_TOKEN | Your Apify API token. Runs are billed to your Apify account at each actor's Store price. | 필수비밀 |
자동 검사
2026년 11월 14일에 실행됨전체 보고서 →자동 검사를 통과했습니다. 이는 보안 감사가 아닙니다. 코드를 직접 읽고 서버에 필요한 최소한의 권한만 부여하십시오. 검사 방식
- 소스 저장소 연결 및 접근 가능subimpact/data-mcp is public and reachablegithub.com/subimpact/data-mcp ↗통과
- 패키지가 이 저장소를 소스로 명시@subimpact/data-mcp names subimpact/data-mcp as sourcenpmjs.com/package/@subimpact/data-mcp ↗통과
- 빌드 출처(Sigstore 증명)No build attestation published for this release검사되지 않음
- 오픈 소스 라이선스 (OSI)MITspdx.org/licenses/MIT.html ↗통과
- 최근 1년 내 커밋 또는 릴리스Last commit or release 2026-10-29github.com/subimpact/data-mcp ↗통과
- OpenSSF Scorecard검사되지 않음
- 패키지 이름이 인기 패키지와 유사하지 않음No popular package one edit away from @subimpact/data-mcp통과
- 패키지 및 해당 종속성에 중요한 보안 권고 없음99 resolved dependencies checked; no critical advisoryosv.dev ↗통과
- 모델에 대한 숨겨진 지침 없음0 tool descriptions and the README read; no override phrases, credential paths or hidden characters통과
- 파괴적인 도구에 라벨 표시됨검사되지 않음
- 원격 엔드포인트가 MCP 핸드셰이크에 응답검사되지 않음
README
GitHub에서 보기@subimpact/data-mcp
Public-data tools for AI agents, in one MCP server. Point it at your own Apify account and your model can search Google (organic results, People Also Ask, knowledge graph and full AI Overviews), run Google AI Mode queries, watch news mentions across Southeast Asian publishers, collect Google Maps places, Google Jobs, Trends, website contacts, page-to-markdown, Ads Transparency data, YouTube, TikTok, Eventbrite, ATS career pages, Substack, Bluesky, Lazada products, JSON-LD entity graphs, SEO audits and Pinterest — all as JSON instead of browser tabs. Every tool wraps one actor from the subimpact Apify Store: the server starts the actor with your APIFY_TOKEN, waits for it, and returns the dataset items. Nothing is scraped locally and no credentials of ours are involved.
Install
You need an Apify API token: console.apify.com → Settings → Integrations. Runs are billed to that Apify account at each actor's Store price (pay per event).
Claude Code
claude mcp add subimpact-data -e APIFY_TOKEN=... -- npx -y @subimpact/data-mcp
Claude Desktop
Add this to claude_desktop_config.json (~/Library/Application Support/Claude/ on macOS, %APPDATA%\Claude\ on Windows):
{
"mcpServers": {
"subimpact-data": {
"command": "npx",
"args": ["-y", "@subimpact/data-mcp"],
"env": {
"APIFY_TOKEN": "apify_api_your_token"
}
}
}
}
Cursor
Add the same block to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):
{
"mcpServers": {
"subimpact-data": {
"command": "npx",
"args": ["-y", "@subimpact/data-mcp"],
"env": {
"APIFY_TOKEN": "apify_api_your_token"
}
}
}
}
Want a smaller tool list? Add "--tools", "google_search,news_mentions" to args, or set SUBIMPACT_TOOLS=google_search,news_mentions in env. get_run_results is always available.
Tools
| Tool | What it does | Price |
|---|---|---|
google_search |
Google search as JSON: full AI Overview answers (lists, citations, sources) plus organic results, People Also Ask, knowledge graph, top stories and local results in one… | Google SERP: $0.003 |
google_ai_mode |
Real Google AI Mode answers (udm=50 tab) rendered in Android Chrome, for any country and language: full answer, structured blocks, resolved source URLs and per-paragraph… | AI Mode search completed: $0.003 + platform usage |
news_mentions |
Search a curated index of publisher news feeds by keyword and get clean article rows: title, snippet, publisher URL, source, region, topic, language and publish time.… | Article: $0.001 + platform usage |
google_maps_places |
Extract Google Maps places with phone, website, rating, reviews, opening hours, popular times, images and place ID. Plain HTTP, no browser, $1.00 per 1,000 places.… | Place scraped: $0.001; Review scraped: $0.0005; Actor start: $0.00005; Add-on: Company contacts: $0.001 + platform usage |
google_jobs |
Scrape Google Jobs listings by query and country: title, company, location, source (LinkedIn, Indeed...), posted date, salary, job type, description and apply links. No… | Job: $0.001 + platform usage |
website_contacts |
Extract emails, phone numbers and social profile links (Facebook, Instagram, LinkedIn, X, TikTok, YouTube, WhatsApp, Telegram) from any list of websites. Plain HTTP, no… | Page scanned: $0.0008 + platform usage |
google_trends |
Scrape Google Trends interest-over-time, related queries/topics, and per-region interest for keywords. No API key, no browser, no captcha. Up to 5 keywords compared per… | Keyword trends bundle: $0.0025 + platform usage |
web_page_markdown |
Web search and fetch tool for AI agents and RAG pipelines. Queries Google Search, scrapes the top N pages over raw HTTP, and returns clean Markdown. $0.001 per page plus… | Page extracted: $0.001 + platform usage |
google_ads_transparency |
Scrape Google Ads Transparency Center: find advertisers by name or ID and export their ads with format, first and last shown dates, preview and ad links, filtered by… | Ad: $0.0008; Advertiser: $0.0005 + platform usage |
youtube |
Fast YouTube scraper with no browser and no API key: keyword search with filters, channel videos and Shorts, channel About data, playlists, video details (exact likes,… | Result: $0.001 + platform usage |
tiktok_hashtag |
High-speed TikTok hashtag scraper. Extract videos, views, likes, comments, shares, audio details, and creator profiles by hashtag at an ultra-competitive price. | Video: $0.001 |
eventbrite_events |
Scrape Eventbrite events with venue and organizer data by default. Search by city, keyword, category, price and dates, or by event URL or organizer. $2.00 per 1,000… | Event: $0.002 + platform usage |
company_career_jobs |
Get live job postings straight from company career pages — Greenhouse, Lever, Ashby, Workday, SmartRecruiters — normalized into one schema, with new-job monitoring for… | Job scraped: $0.001; Company monitored: $0.02 + platform usage |
substack |
Substack Intel API — Posts, Archives & Publication Stats | Run started: $0.00005; Post or publication row: $0.001 + platform usage |
bluesky |
Scrape Bluesky posts, profiles, threads, and follower graphs by keyword, hashtag, or handle. No login required. Fast AT Protocol API, JSON/CSV export, pay only per… | Post scraped: $0.001; Profile scraped: $0.002 + platform usage |
lazada_products |
Scrape Lazada search results and category pages across 6 SEA markets (MY/SG/TH/ID/PH/VN) with a real browser session via Apify residential proxy. Monitor price drops,… | Product scraped: $0.00099; Product checked: $0.003; Alert created: $0.01 + platform usage |
jsonld_entities |
Extract deeply nested JSON-LD, validate against Schema.org (979 types, 1,633 properties), and resolve disconnected blocks into one unified knowledge graph with typed… | Actor start: $0.00005; Page processed: $0.0015 + platform usage |
seo_page_audit |
Port of the SEO Minion extension as an API: full on-page audits per URL (meta tags, headings, images, links, structured data, hreflang) plus Google SERP snippet previews… | page audit: $0.001; SERP preview: $0.001 + platform usage |
pinterest |
Scrape Pinterest pins from keyword search, boards and profiles. Full-res originals, video variants, board/pinner metadata, repin counts. Pure HTTP - no browser, no… | Pin scraped: $0.00199; Media downloaded: $0.005 + platform usage |
rss_feed_reader |
Read RSS, RDF, Atom and JSON Feed URLs and get one clean JSON row per item: title, link, ISO dates, author, summary, categories, image, enclosures. Only-new mode for… | Feed item: $0.0005 + platform usage |
telegram_channel |
Scrape public Telegram channels via the t.me web preview: one row per post with text, date, views, media, link previews, polls, reactions, hashtags and links, plus… | Post: $0.0003; Channel info: $0.001 + platform usage |
google_flights |
Search Google Flights and get one row per flight option: fare, airlines, flight numbers, times, airports, stops, layovers, aircraft, legroom and CO2. One-way or round… | Flight search: $0.002 + platform usage |
my_grocery_prices |
One normalised Malaysian grocery price feed from 99 Bulksales, Jaya Grocer and KPDN PriceCatcher, with pack-size parsing, RM per kg / litre / piece unit prices and… | Product price row: $0.001 + platform usage |
Every tool also accepts two optional control parameters:
_maxItems(integer, default 50, max 1000) — how many dataset items come back to the model. It truncates the response only; the actor still collects (and charges for) what its input asks for._maxChargeUsd(number, default 0.5) — per-run spending cap in USD, sent to Apify as the run optionmaxTotalChargeUsd. The run stops when it reaches the cap, so an accidental large batch cannot cost more than this.
The extra get_run_results tool (runId, _maxItems, offset) fetches the status and items of a run that was still running when its tool call returned. Long jobs return a runId instead of blocking forever; call it later to collect the results.
Cost control
Every call is a real Apify run on your account, priced per event by the actor (for example google_search at $0.003 per SERP, news_mentions at $0.001 per article). There is no subscription and this server never charges you: Apify bills your account directly.
- Start with the default
_maxChargeUsdof$0.50; it is passed to Apify and enforced by the platform. - Use
_maxItemsto keep model context (and token spend on the model side) small. - Narrow the actor input instead of raising the cap when you need fewer, more targeted results.
- Each response includes
itemCount(total items in the dataset) andreturned(items sent to the model), so you can see when a result was truncated.
A single tool call waits up to SUBIMPACT_TIMEOUT_SECS (default 240 seconds). If the run is still going, you get its runId and can collect the results with get_run_results.
Configuration
| Variable | Required | Description |
|---|---|---|
APIFY_TOKEN |
yes (for calls) | Your Apify API token. tools/list works without it so directories can introspect the server. |
SUBIMPACT_TOOLS |
no | Comma-separated subset of tool names to expose, e.g. google_search,news_mentions. |
SUBIMPACT_TIMEOUT_SECS |
no | Per-call wait budget in seconds (default 240). |
SUBIMPACT_API_BASE |
no | Override the Apify API base URL (default https://api.apify.com/v2); meant for testing against a mock. |
CLI flags: --tools a,b,c, --help, --version.
What it is not
- Not a local scraper. It does not fetch pages or parse HTML itself; it runs Apify actors, which do the scraping on Apify's infrastructure.
- Not a credential holder of ours. The only token it uses is the
APIFY_TOKENyou set. The server ships with no secrets and calls no private backend. - Not free. Every tool call is an Apify run billed to your Apify account at the actor's Store price. The
_maxChargeUsdcap limits a single run, not your monthly Apify spend. - Not an official Apify product, and not affiliated with Google, TikTok, Meta, Lazada or the other platforms whose public data the actors read.
- Not a guarantee of data. Public sites change; review each actor's Store page for current coverage, pricing and limitations.
Development
npm install
npm run build # tsc + copy src/tools.json into dist/
npm test # offline tests: mocked fetch + a stdio round trip
npm run typecheck
npm run gen:readme # regenerate the tool table above from src/tools.json
node dist/index.js # run the stdio server
src/tools.json is a verbatim snapshot of public Apify Store actor metadata: tool names, descriptions, pay-per-event pricing and input schemas. The MCP tool definitions, descriptions and this README's tool table are generated from it.
License
MIT © 2026 subimpact network
subimpact data MCP 질문
subimpact data MCP를 Claude Code에 추가하려면 어떻게 하나요?
터미널에서 claude mcp add --env 'APIFY_TOKEN=<YOUR_APIFY_TOKEN>' data-mcp -- npx -y @subimpact/data-mcp를 실행합니다. 모든 프로젝트에서 사용할 수 있게 하려면 --scope user를 추가합니다.
subimpact data MCP를 Cursor에 추가하려면 어떻게 하나요?
~/.cursor/mcp.json를 열고 mcpServers 아래에 서버를 추가합니다. 위 설치 섹션에 표시된 Cursor 구성을 사용합니다.
이 도구들을 사용하는 데 비용이 얼마나 드나요?
비용은 각 actor의 Store 이벤트당 가격에 따라 Apify 계정으로 직접 청구됩니다.
도구 호출 한 번당 발생하는 비용을 제한할 수 있나요?
네, _maxChargeUsd 파라미터를 사용하여 실행당 USD 기준 지출 한도를 설정할 수 있습니다.
도구 실행 시간이 길어지면 어떻게 되나요?
작업이 오래 걸리는 경우 도구가 runId를 반환하며, 이후 get_run_results 도구를 사용하여 데이터를 수집할 수 있습니다.