Finance Toolkit
500+ transparent financial methods calculated from raw statements, not third-party endpoints.
제작자가 쓴 설명(영어 원문)
Indian stocks: NSE/BSE filings read by AI with price reactions, fundamentals, IPOs and FPI flows.
이 세 가지는 절대 하나의 점수로 합치지 않아요. 의견, 측정, 인기는 서로 다른 것이니까요.
이 서버는 아직 쉬운 말로 된 요약을 쓰지 못했어요. 아래는 레지스트리에서 읽어 온 사실이에요.
쓰고 있는 AI 앱을 고른 다음 설정을 복사하세요.
필요한 것
어떤 앱을 쓰세요?
위에서 앱을 고르세요. 기억해 두었다가 다음부터는 버튼에 “…에 설치”라고 표시해요.
웹, 데스크톱, 모바일의 Claude에서 쓸 수 있어요:
https://mcp.marketping.in/mcp터미널에서 이 명령어를 실행하세요
claude mcp add --transport http marketping https://mcp.marketping.in/mcp프로젝트의 .cursor/mcp.json 또는 ~/.cursor/mcp.json에 붙여 넣으세요
{
"mcpServers": {
"marketping": {
"url": "https://mcp.marketping.in/mcp"
}
}
}브라우저가 앱을 열지 물어본 다음, 앱이 설정을 보여 주고 확인을 요청해요. 키 값은 비워 두니 직접 입력하세요.
프로젝트의 .vscode/mcp.json에 붙여 넣으세요
{
"servers": {
"marketping": {
"type": "http",
"url": "https://mcp.marketping.in/mcp"
}
}
}브라우저가 앱을 열지 물어본 다음, 앱이 설정을 보여 주고 확인을 요청해요. 키 값은 비워 두니 직접 입력하세요.
ChatGPT는 인터넷에 있는 서버에만 연결할 수 있어요:
https://mcp.marketping.in/mcp메뉴 이름은 요금제나 시기에 따라 조금 다를 수 있어요.
선택한 앱은 이 기기의 쿠키에만 저장돼요. 계정은 필요 없어요.
이 설정은 제작자의 레지스트리 정보로 만들었어요. README와 다르면 README를 따르세요.
도구 이름과 설명은 서버가 2026년 11월 9일에 알려 준 그대로예요. 우리가 아니라 제작자가 쓴 글이고, 일반 텍스트로만 보여 드려요.
searchSearch MarketPing's Indian stock market data — companies, corporate announcements, IPO issues, and the text of annual reports, earnings-call transcripts and investor presentations. Returns results with ids to pass to `fetch`. Factual data only — MarketPing publishes no buy/sell r
fetchRetrieve the full content of one search result by its id — a company profile with fundamentals and recent announcements, one announcement with its measured price reaction, one page of a filing, or one IPO issue. Factual data only — MarketPing publishes no buy/sell recommendations
resolve_companyFind an Indian listed company by name, ticker or partial match. Returns candidates with their MarketPing slug — call this FIRST unless you already hold a slug, because every other company tool takes one. Factual data only — MarketPing publishes no buy/sell recommendations, no pri
company_snapshotLatest fundamentals and valuation for one company: price, market cap, P/E, P/B, TTM revenue and profit, growth, ROCE/ROE and debt ratios. Use for 'how is X doing', 'what does X trade at', or as the numeric backdrop before reading filings. Factual data only — MarketPing publishes
recent_alertsRecent NSE/BSE corporate filings with MarketPing's AI classification — subject, category, document type, importance and sentiment. This is MarketPing's core dataset: use it for 'what has X announced lately', 'any news on X', or (with no company) 'what is happening in the market t
alert_reactionHow the market actually moved after ONE announcement: percentage moves at 5 and 30 minutes from the announcement-minute price, and on the reaction day (the first session that could trade the news), the next day, 2/3/5/7/15 sessions and 1/2/3 months from the close before the filin
search_filingsFull-text search across every page MarketPing holds — annual reports, earnings-call transcripts, investor presentations. Returns {doc_id, page, snippet} hits to follow up with read_pages. SNIPPETS ARE KEYWORD WINDOWS, NOT FACTS: never state a figure from a snippet, open the page
list_documentsThe documents MarketPing holds for a company — annual reports, earnings-call transcripts, investor presentations, results — newest first, each with a doc_id. A capped list is marked as truncated: never read one as the whole holding. Factual data only — MarketPing publishes no buy
doc_contentsA document's table of contents: section titles, one-line summaries and page ranges. Read this BEFORE read_pages so you open the right four pages instead of guessing — annual reports run to 300+ pages. Factual data only — MarketPing publishes no buy/sell recommendations, no price
read_pagesThe actual text of a page range (max 12 pages). This is the ONLY source you may quote a document from — cite as {doc_id, page, verbatim quote}, and the quote must really appear on that page. Factual data only — MarketPing publishes no buy/sell recommendations, no price targets an
audit_reportAuditor opinions, key audit matters, CARO exceptions and auditor changes per fiscal year, extracted page-cited from annual reports. Use for governance and red-flag questions. A missing year means not yet read — never read it as 'clean'. Factual data only — MarketPing publishes no
screener_fieldsThe metrics you can filter and sort on in run_screener, with labels, units and groups. Call this before building a screen if you are unsure of a field name. Factual data only — MarketPing publishes no buy/sell recommendations, no price targets and no IPO grey-market premium.
run_screenerFilter MarketPing's universe of 5,400+ NSE/BSE companies on fundamentals — 108 metrics covering valuation, growth, profitability, leverage and ownership. Use for 'find companies where…' questions. Field names come from screener_fields. Factual data only — MarketPing publishes no
ipo_pipelineIndian IPOs — open now, upcoming, closed or already listed — with price band, issue size, lot size, dates, subscription figures and post-listing performance. Covers mainboard and SME. MarketPing deliberately publishes no grey-market premium and no apply/avoid verdict; if you need
fpi_flowsForeign portfolio investor (FPI/FII) flows into Indian markets as NSDL publishes them — daily net equity and debt flows, by investment route, with month-to-date totals. Use for 'are foreign investors buying' questions. Factual data only — MarketPing publishes no buy/sell recommen
results_calendarWhich Indian companies report quarterly results on which dates, as the exchanges have published them. Default window is the next 60 days. Factual data only — MarketPing publishes no buy/sell recommendations, no price targets and no IPO grey-market premium.
daily_briefingMarketPing's briefing for the latest NSE trading day: market score, index and sector moves, FII/DII and FPI flows, breadth, movers, and the day's most important announcements. Use for 'how did the Indian market do today'. Factual data only — MarketPing publishes no buy/sell recom
list_stacksMarketPing's curated stacks: public lists of Indian companies rebuilt every day from layers (quality/value/growth screens, an index, a Discover catalyst page) with one rule — how many layers a company must clear today. Returns each stack's slug, rule, member count and what entere
run_stackThe companies in one stack as of today's snapshot, with how many layers each clears, which ones, how many consecutive days it has been in the stack, market cap and P/E. Takes a slug from list_stacks. Membership is a fact about screens cleared today — never a rating. Factual data
stack_changesWhat entered, left, cleared more or fewer layers in a stack since a date (default: the previous daily snapshot). Use for 'what is new in this list today / this week'. Factual data only — MarketPing publishes no buy/sell recommendations, no price targets and no IPO grey-market pre
list_modelsThe valuation models saved to this MarketPing account (Models: DCF, excess return, sum of parts), with the company and a link. Needs a linked account. Optional `company` narrows to one company (slug). Factual data only — MarketPing publishes no buy/sell recommendations, no price
run_modelRecompute one of the account holder's saved models (DCF, excess return or sum of parts) with MarketPing's engine and return the value per share, the bridge and the projected years (3, 5 or 10, as the model sets). The figure is the holder's own arithmetic over their inputs — repor
watchlist_listThe companies this MarketPing account follows, with their alert status. Needs a linked MarketPing account — the assistant will offer to connect one. Factual data only — MarketPing publishes no buy/sell recommendations, no price targets and no IPO grey-market premium.
watchlist_addAdd a company to this MarketPing account's watchlist, so its filings start arriving as alerts. Needs a linked MarketPing account. Subject to the plan's company limit. Factual data only — MarketPing publishes no buy/sell recommendations, no price targets and no IPO grey-market pre
누구나 볼 수 있는 정보를 점검했고, 코드는 읽지 않았어요. 해당하지 않는 항목은 개수에서 뺐어요. · 2026년 11월 9일 점검
레지스트리 이름이 도메인 marketping.in 기준으로 확인되었어요(레지스트리가 도메인 소유를 확인해요)
아직 GitHub에서 이 저장소를 읽어 오지 않았어요
아직 GitHub에서 이 저장소를 읽어 오지 않았어요
836ms 만에 응답(HTTP 200)
로그인 필요 없음
도구 설명 24개를 읽었고, AI에게 보내는 숨겨진 지시는 없었어요
레지스트리에서 이 카테고리의 활성 서버 전체와 비교했어요. 스타는 게시자가 저장소를 소유한 경우에만 세요.
해당하는 기본 점검 3개 중 3개를 통과했어요. 코드는 읽지 않았어요. 키나 토큰을 넣기 전에 서버가 어떤 권한을 요구하는지 확인하고, 꼭 필요한 권한만 있는 키를 쓰세요.
레지스트리에 필수 키가 나와 있지 않아요. 다만 처음 쓸 때 로그인을 요구하는 서버도 있어요.
제작자의 클라우드에서 실행돼요. 설치 없이 인터넷으로 연결하지만, 보내는 내용은 제작자의 서버를 거쳐요.
GitHub 저장소는 마지막으로 – 변경됐고, 최신 버전(1.0.1)은 2개월 전 레지스트리에 올라왔어요.
Claude Desktop, Claude Code, Cursor, VS Code처럼 MCP를 지원하는 앱에서 쓸 수 있어요. URL로 연결되기 때문에, 원격 MCP 서버를 지원하는 앱(예: 개발자 모드의 ChatGPT)에서도 쓸 수 있어요.
같은 카테고리에는: Finance Toolkit, SolidInvoice, Korea Tax Calculator
같은 카테고리에서 다른 제작자가 만든 서버예요. 스타 순으로 정렬했어요.
카드 보는 법:↓ 주간 다운로드i★ GitHub 스타i세 기둥: 리뷰 · 점검 · 사용 정도i업데이트 = 최신 버전을 레지스트리에 공개한 날i
500+ transparent financial methods calculated from raw statements, not third-party endpoints.
Create and send invoices and quotes, manage clients, and track payments in SolidInvoice.
Korean tax calculation and tax-audit judgment MCP: 96 tools with statutory citations.
제작자의 저장소에서 쓰인 그대로 가져왔어요. PopNx MCP가 검토하거나 보증한 내용이 아니에요. HTML, 이미지, 스크립트는 제거했어요. 링크를 누르면 이 사이트를 벗어나요.
Indian stock-market research for AI assistants: NSE/BSE corporate filings read by AI with the measured price reaction after each one, fundamentals for 5,000+ companies, page-cited reads of annual reports and earnings calls, a 108-metric screener, the IPO pipeline, FPI flows, the results calendar and the daily market briefing.
https://mcp.marketping.in/mcp — remote, Streamable HTTP, JSON-RPC 2.0, MCP protocol 2025-06-18watchlist_add), annotated readOnlyHint: false. Everything else is read-only.This repository is the server's public contract: the exact tool catalogue the server returns from tools/list (tools.json), a registry manifest (server.json), and a small open-source stdio bridge for clients that cannot speak Streamable HTTP directly. The server itself runs on MarketPing's infrastructure; its handlers are thin read-only wrappers over the same code paths that serve marketping.in.
Claude (claude.ai / Claude Desktop) — Settings › Connectors › Add custom connector
https://mcp.marketping.in/mcp
ChatGPT (Developer mode / Apps) — add MCP server: same URL.
Cursor / VS Code / Windsurf (`mcp.json`):
{
"mcpServers": { "marketping": { "url": "https://mcp.marketping.in/mcp" } } }
Clients that only speak stdio — bridge to the same endpoint with mcp-remote:
{
"mcpServers": { "marketping": { "command": "npx", "args": ["-y", "mcp-remote", "https://mcp.marketping.in/mcp"] } } }
That runs the published mcp-remote bridge against the endpoint above and nothing else. It stores no credentials of its own; mcp-remote keeps OAuth tokens in its standard location (~/.mcp-auth) only if you link an account.
bin/marketping-mcp.js in this repo does the same thing as a named wrapper. It is not published to npm, so npx -y @marketping/mcp will not resolve — run it from a clone, or just use the mcp-remote form above.
Exact schemas and annotations: tools.json (the live tools/list response, captured 19 Sep 2026). Every description ends with the no-advice line — that is part of the contract the consuming model reads.
| Tool | What it returns | Auth |
|---|---|---|
search | Companies, filings, IPOs and documents matching a query (ChatGPT search contract) | none |
fetch | The full content of one search result by id | none |
resolve_company | A listed company by name/ticker/partial match, with candidates | none |
company_snapshot | Latest fundamentals and valuation for one company | none |
recent_alerts | Recent NSE/BSE filings with AI subject, category, importance, sentiment | none |
alert_reaction | The measured move after ONE filing: 5 min … 3 months | none |
search_filings | Full-text search across every document page MarketPing holds | none |
list_documents | The documents held for a company, newest first | none |
doc_contents | A document's table of contents with page ranges | none |
read_pages | The text of a page range (max 12 pages) — the only quotable source | none |
audit_report | Auditor opinions, key audit matters, CARO exceptions per fiscal year | none |
screener_fields | The 108 metrics available to run_screener | none |
run_screener | Filter the 5,000+ company universe on fundamentals | none |
ipo_pipeline | Open / upcoming / closed / listed IPOs with issue data and subscription | none |
fpi_flows | FPI/FII flows as NSDL publishes them | none |
results_calendar | Who reports results on which dates | none |
daily_briefing | The latest trading day's briefing: score, indices, sectors, flows, movers | none |
list_stacks | MarketPing's curated stacks (public daily-rebuilt company lists) | none |
run_stack | The members of one stack as of today's snapshot | none |
stack_changes | What entered/left a stack since a date | none |
list_models | The valuation models saved to the linked account | linked account |
run_model | Recompute one saved model with overrides | linked account |
watchlist_list | The companies the linked account follows | linked account |
watchlist_add | Write. Add a company to the linked account's watchlist | linked account |
mcp.marketping.in over HTTPS. Nothing else leaves the client; the bridge adds no telemetry.https://mcp.marketping.in/.well-known/oauth-authorization-server and /.well-known/oauth-protected-resource) unlocks the four account tools and the account's plan budget (Investor 1,000 / Trader 2,000 calls per day). Tokens are issued by MarketPing's own authorization server and can be revoked from the account's settings.marketping.in URL so the assistant can cite the page. Document text comes only from read_pages, page-anchored, so quotes are checkable.curl -s -X POST https://mcp.marketping.in/mcp \
-H "Content-Type: application/json" -H "Accept: application/json, text/event-stream" \ -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"curl","version":"1"}}}'
curl -s -X POST https://mcp.marketping.in/mcp \ -H "Content-Type: application/json" -H "Accept: application/json, text/event-stream" \ -d '{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}'
The second response is tools.json.
MarketPing is an Indian stock-market research platform built around one idea: every NSE and BSE corporate filing, read by AI the moment it is published, delivered to WhatsApp within a minute, and followed by the measured price reaction — what the stock actually did 5 minutes to 3 months later. 500,000+ filings decoded, 1.9 million reactions measured, 5,000+ companies. It publishes no buy/sell calls, no price targets and no ratings.
What the same data looks like on the site (every MCP result links back to one of these):
The bridge and the files in this repository are MIT-licensed (see LICENSE). The MarketPing service and its data are proprietary; use of the endpoint is subject to https://marketping.in/terms.