Advertise
MCP.so
Sign In

Wealthnow MCP ServerVerifiedFeatured

@Wealthnow

RepositoryHomepageDocumentation

Basic information

Category

Finance & Commerce

License

MIT

Auth

oauth

Author

Wealthnow

Submitted by

Humberto Lobo

About Wealthnow MCP Server

351 read-only tools on institutional data: stocks since 1926, vol surfaces since 1996, CDS, bonds, private markets, 13F, macro. One OAuth connection.

#finance#market-data#stocks#options#credit#private-markets#sec-filings#crypto#macro#research

Wealthnow MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "wealthnow": {
      "url": "https://mcp.wealthnow.io/mcp"
    }
  }
}

Wealthnow MCP Server tools

40

Your Wealthnow account: its plan, the balance it has left and its entitlements. Needs your connected account or your API key (X-API-Key). Reading it does not draw on the balance. The balance is read from the key service when you call: as_of is when it was read, and stale_seconds how long ago (0 unless the key service did not answer).

Find tools in the whole Wealthnow catalogue by topic, including tools not in your tool list; query * lists every tool, in name order. Returns one page of matches, total_matches, and next_offset (pass it as offset for the next page; null on the last). Each match has its name, a one-line summary, its inputSchema, in_tool_list (whether it is in your tool list) and, when your account can be read, available_to_you: whether your account can use the tool. Run a match that is not in your tool list with call_tool. It only searches the catalogue; it runs no tool.

Run a Wealthnow tool that search_tools found and that is not in your tool list. Pass its exact name and its arguments, as its inputSchema says. A tool that records data is not run here: call it directly.

Quotes, charts, ETF holdings and crypto market feeds (funding, open interest, basis, on-chain flows). One tool for 13 operations: set operation to one of them and pass that operation's arguments in arguments.

Company fundamentals: financial statements, metrics, growth, dividends, peers, screener, XBRL facts and price snapshots. One tool for 22 operations: set operation to one of them and pass that operation's arguments in arguments.

SEC filings and filing section text. One tool for 2 operations: set operation to one of them and pass that operation's arguments in arguments.

Congressional trades, politicians and corporate political donors. One tool for 3 operations: set operation to one of them and pass that operation's arguments in arguments.

Equity, crypto and market news, sentiment, analyst ratings, press releases and trending tickers. One tool for 43 operations: set operation to one of them and pass that operation's arguments in arguments.

Yield curves, economic and earnings calendars, commodities, COT, forex news and market regime. One tool for 26 operations: set operation to one of them and pass that operation's arguments in arguments.

ML predictions and drivers, factor exposures, signal fusion, regime forecasts and backtest statistics. One tool for 39 operations: set operation to one of them and pass that operation's arguments in arguments.

13F holdings, insider trades, institutional and mutual-fund owners, estimates, price history and corporate actions. One tool for 19 operations: set operation to one of them and pass that operation's arguments in arguments.

Coins, prices, charts, exchanges, derivatives exchanges, on-chain pools and tokens, and trending. One tool for 35 operations: set operation to one of them and pass that operation's arguments in arguments.

Insider and ownership filings, government contracts, lobbying, patents, workforce, supply chain, point-in-time fundamentals, macro history and identifiers. One tool for 43 operations: set operation to one of them and pass that operation's arguments in arguments.

Options flow, gamma exposure, dark pools, short interest, borrow rates and options chains. One tool for 29 operations: set operation to one of them and pass that operation's arguments in arguments.

Equity briefs, technical analysis, trading plans, TWAP and VWAP schedules and scenario simulation. One tool for 8 operations: set operation to one of them and pass that operation's arguments in arguments.

Web search, page fetch, research synthesis, framework lookup and AI fundamentals analysis. One tool for 5 operations: set operation to one of them and pass that operation's arguments in arguments.

Private companies, deals, investors, funds, limited partners, valuations and relations. One tool for 28 operations: set operation to one of them and pass that operation's arguments in arguments.

Equity, index, futures and options bars, chains and microstructure. One tool for 9 operations: set operation to one of them and pass that operation's arguments in arguments.

Bonds, loans, CDS history, credit indices and ratings. One tool for 6 operations: set operation to one of them and pass that operation's arguments in arguments.

Corporate events, earnings-call transcripts, conference calls and the IPO pipeline. One tool for 8 operations: set operation to one of them and pass that operation's arguments in arguments.

VIX indices, VX futures curve and COT positioning. One tool for 3 operations: set operation to one of them and pass that operation's arguments in arguments.

Crypto derivatives, venues, on-chain metrics, events and volatility surfaces over time. One tool for 5 operations: set operation to one of them and pass that operation's arguments in arguments.

Index hedge proposals (SPY/QQQ/IWM) computed from live dealer positioning — protective puts when dealers are short gamma and IV rank is cheap, plus an IV term-structure note when front-week IV is at least 5% above 30-day IV. No collar is proposed: 'withheld_rules' says why. 'units' says what each metric is measured in

Real-time X/Twitter sentiment narrative. Pass ticker=NVDA for a focused fintwit read on a name, or query=... for a free-form social-media question. Returns a model-written narrative (is_model_output true) and any source URLs it grounded against; grounded says whether the model searched X. There is no structured bullish/bearish ratio, and prices, moves and counts in the text are not checked against market data: quote numbers from the market-data tools. Use when you want the *vibe* on a name right now (retail sentiment, breaking rumours, unusual social activity), not the news article list. When the provider does not answer it returns 503 upstream_unavailable (error_code social_sentiment_unavailable; not billed). Each answer runs a live grounded search, and a repeat of the same ticker or query within 5 minutes is served from that answer.

Search PRIVATE companies / investors (VC/PE) / funds / people / limited partners by name (prefix, case-insensitive), ticker, or CIK — relevance-ranked so the prominent entity is #1; an exact ticker or CIK leads. Matches also-known-as, legal and former names and the homepage domain ('Nubank' finds Nu Holdings through nubank.com.br). Use this FIRST for any private-company question (e.g. 'tell me about Stripe', 'who is Sequoia') to resolve the entity id, then call the company/dossier/realtime tools. type=all searches every entity kind. detail=full returns every column per hit (for rich tables).

The private-company DETAIL PAGE in ONE call, render-ready: identity + key facts + the valuation/revenue/headcount tapes + the financing-in-progress card + the team roster. Money ships as both a raw `*_musd` float and a formatted `*_display` string, every series is sorted ASCENDING for charting, and the hero valuation badge is computed server-side (never presented as an estimate unless it is one). Blocks degrade independently. Use this for a company PAGE; use /dossier for the analytical fan-out (investors/competitors/similar) and /realtime for live overlay.

Funding-round and M&A deal history for a private company (deal size, type, VC round, pre/post-money valuation), newest first. Call this when the user asks 'when did X last raise / at what valuation / who acquired it'; use /investors for who participated.

One-shot fundamentals bundle for a ticker, fetched in parallel: the metrics snapshot (same blocks and labels as fundamentals_metrics_snapshot), statements in MATCHED periods (income_statement_ttm with cash_flow_ttm, income_statement_latest_quarter with cash_flow_latest, balance_sheet_latest at the quarter end; `period_pairs` says which match) and company facts. Every block carries `period` and `units`. Cash-flow panes reconcile: the fields in `operating_cash_flow_components` sum to operating_cash_flow; change_in_working_capital is null with a status because the source's figure is not working capital. Insider trades (the last 5 rows, or the newest Form 4 whole when it has more lines, with insider_trades_note) and the top 10 institutional holders (13F, with report period and freshness) need a plan with the funds product; without it they are null with status plan_required. A null pane always has a status; [] means none reported. PRIMARY tool for 'give me the full fundamental picture of X' — call it instead of assembling the pieces one by one. Snapshot statement blocks carry statement_check.

Latest quote for one ticker. Call this when the user asks what the price is now or how the stock is moving today; for history use /fundamentals/prices. price is the last trade in ANY session; price_session says which (the closing auction print, stamped just after the bell, is regular) and as_of when. Sessions, US Eastern: pre 04:00-09:30, regular 09:30-16:00, post 16:00-20:00 (half-days: regular to 13:00, post to 17:00), closed otherwise and on weekends and exchange holidays. market_session (and session, the same value) is the session when the answer is served. day_change = price - prev_close, and day_change_percent is that over prev_close in percent points (1.5 = 1.5%); prev_close is the regular-session close of prev_close_date. Outside regular hours they include extended-hours trading. Every change is the exact price difference and every percent has 6 significant figures. For how the stock did in its latest regular session use regular_session {date, status in_progress|closed, open, high, low, close (null until final, 5 minutes after the bell), last, volume (whole shares, extended hours included: once the session is final, the daily aggregate's, the figure /fundamentals/prices serves; while it forms, the live day bar's running tally, which can differ by a few hundred shares), prev_close, prev_close_date, change, change_percent}; for the move since that close use extended_hours {session pre|post, price, as_of, reference_close, reference_close_date, change, change_percent}, null in regular hours or when the last trade was a regular-session trade. day_* is the bar of day_bar_date (before the open, the previous session's); day_volume is its running tally in whole shares. bid/ask are served only while the exchange is in a session, and only a quote set during that session (quote_as_of, when the quote last changed, at or after the session's start); else null. session_detail_status says why a session field is null: partial (a date or an earlier close could not be established) or unavailable. State the session and date with any move; never report an extended-hours move as a session's change. CRYPTO: pass asset_class=crypto for BTC/ETH/SOL/LTC/LINK etc. Several crypto symbols are ALSO US-listed equity tickers (BTC is a Grayscale trust at ~$29; LINK is Interlink Electronics), so a bare ticker returns the EQUITY. Never use an equity price for a crypto asset. Check is_stale before using the price. VALUATION (equity, on the live price, when valuation_status is 'ok'; otherwise null): pe_ttm = price / eps_ttm (diluted, four consecutive fiscal quarters, today's share basis); forward_pe = pe_ttm / (1 + the latest fiscal year's EPS growth, or net-income growth when a split sits between the years), trailing arithmetic, NOT an analyst consensus, and null when that growth is zero or negative (forward_pe_status says why); market_cap = price x market_cap_shares, the SEC cover page's shares outstanding when current, else the reference count, else the latest balance sheet's, else the latest quarter's weighted-average basic count (market_cap_shares_basis, market_cap_shares_as_of), null for a listing that is not the issuer's equity (a note, preferred, warrant or ETN), a listing without a share count of its own, and when the split history cannot be read now (market_cap_shares_status says why or when it is provisional); dividend_yield is a FRACTION (0.0225 = 2.25%): split-adjusted regular cash dividends with an ex-date in the last 365 days over the price, null for a non-payer; dividend_yield_indicated uses the latest declared regular dividend x payments per year; `dividend` gives the per-share figures, dates and status; units names each unit.

Unusual options-flow alerts for one ticker from the options-flow feed (default 25), newest first; executed_at is the alert's last print (executed_at_basis says when only the later alert time was available). An unknown symbol is a 404. Call this when the user asks 'any unusual options activity in X?' or wants the large options bets hitting a specific name; use intel_options_flow for the cross-market view.

13F position snapshots from the institutional 13F archive: fund, ticker, shares, value_usd, report_period — the institutional-ownership signal. Call this when the user asks 'which funds hold X?' or 'what does fund Y own?'; filter by ticker and/or fund (substring match). The archive ends a quarter behind EDGAR: answers carry newest_report_period, vendor_ceiling, expected_report_period, is_stale and quarters_behind. A ticker is read on its share class's CUSIP (security_key; GOOGL and GOOG are separate). missing_known_filers and coverage_note name manager families whose 13F filings the archive does not carry for the quarter (Vanguard's successor filers from 2026-03-31), so their positions are absent, not sold. fund is the archive's manager label; fund_current_name gives today's name where the label is a former one (manager_name_note). For a ticker, current_quarter adds the newer EDGAR 13F filings of the tracked managers only (coverage 'tracked_managers_only', report_period on each item); it is not the full holder list. current_quarter_error says those filings could not be read. Alternative-data rows served because the archive could not be read carry archive_error; fallback rows are the filings it has indexed, not a full holder list (coverage_note). When neither source can answer the route returns 503 upstream_unavailable (error_code archive_read_timeout, archive_read_failed or upstream_dead_no_archive; not billed), never an empty list.

Congressional trades from two coverage sources — a realtime cross-ticker feed (provider=options_flow, default) or a warehouse archive (provider=alternative_data), never a live alternative-data fallback. Without a ticker the archive covers the last14 disclosure days (window_days/window_basis); — with an optional ticker filter, each trade with the member's party (null when unknown). With provider=alternative_data and a ticker, trades come from a warehouse copy that holds trades from 2025-05-20 onward (earlier history for 114 tickers); data_from, history_floor and coverage_note say where the copy begins, so never call it the complete history. Both feeds add side (buy/sell/exchange), party_code, chamber_code and is_member_of_congress (false for an executive-branch filer), and owner (whose account: self, spouse, joint, child) where the feed states it. On options_flow each trade also carries asset_type (stock, option, government_security, ...) from the disclosure's asset code, and an option row's side is the side of the option trade (side_applies_to option_contract; option_type, strike and expiry only where the source states them); a House member's own account reads owner self (owner_basis). reported_date_basis says what reported_date is: on options_flow it is the day the source recorded the disclosure, which can trail the filing date by one to three days. Call this when the user asks what Congress members have been buying or selling, market-wide or in a specific name. A 503 means the feed could not be read, not that nobody traded; a ticker the security master does not carry is a 404 unknown_ticker.

One-shot news intelligence for a ticker: stories of the last 24 hours (news_count_24h counts them all; the newest 20 are listed), 7-day sentiment stats, trending stories of the last 72 hours, structured events, and analyst actions in a single parallel fetch (90s TTL). THE primary tool for 'what's the news on X?' — this ALONE answers most single-ticker news questions; do NOT stack other news tools in the same round unless it returns nothing useful. analyst_actions carries only part of the street's rating actions (analyst_actions_note): an empty list is none in the feed, not none taken.

One-call credit snapshot for a company — call FIRST for any 'how risky is this company's debt?' question: latest securities-finance 5Y CDS spread (bps) with ~90-quote trend + market-implied default probability, the current S&P rating of the issuer's OUTSTANDING rated issues (issue grain: withdrawn and matured issues excluded, rating_basis says how; null when no outstanding issue is rated) with the last upgrade/downgrade (last_action; a new issue's first rating is newest_rating_action), a recent FINRA TRACE bond yield/volume summary, and the syndicated-loan-facility count (all-time, plus the facilities still active summed per currency, withheld when the record is historical; data_through dates it). default_prob is the 5-year cumulative risk-neutral probability; the CDS block carries age_days against the archive's end. Blocks degrade independently (a missing dataset returns an error field in its block; a block not answered inside the time budget is skipped_budget); when no block answers, the call is a 503 credit_snapshot_timeout. A symbol neither the security master nor the market-data listing knows is a 404 unknown_ticker. Drill down with the credit bonds/cds/ratings/loans tools.

Search earnings-call transcripts. scope=headlines (default) searches call titles across ALL companies; scope=text searches the spoken words inside one company's transcripts (ticker REQUIRED) and returns per-call hit counts + a snippet. Use when the user asks 'which calls mention X' — then fetch the full text with the returned event_id. Headline results name each call's company. Text results are one per call (best copy), each with its date and title; hits counts the speaker turns that mention the term and the snippet is cut around the first match.

The SUPPLY-CHAIN GRAPH around one company — its customers, suppliers, competitors and partners in one call, each with the relationship's start date, whether it is still open, which side reported it, and revenue dependence where it was estimated. Both directions are merged and normalised to the queried company's point of view, so 'customers' includes companies that report THIS company as their supplier. Call it to map second-order exposure (whose earnings move when this name moves) or to find the listed suppliers behind a product cycle. Defaults to CURRENT relationships; status=all or as_of=YYYY-MM-DD gives history. One entry per counterparty by default (several records can back one pair) — group_by=record gives the underlying versions. revenue_percent belongs to the company named in revenue_percent_of_ticker, is an ESTIMATE on every row, and is present on only a minority of edges; coverage is reported per group.

Use when: the user asks for a specific ticker's next earnings date, when a company reports, the earnings calendar entry for a name, or anything of the form "when is X's next earnings?". This is the CANONICAL multi-source consensus tool — fans out to the company's own SEC 8-K filings (Items 2.02/7.01/8.01, last 45 days), the market-data listing, the estimates calendar, the options-flow calendar, the news feed and web search; reconciles via primacy-weighted majority; returns a single canonical answer with per-source breakdown and deduplicated citations. ``confirmed`` / next_earnings_date mean a source says the COMPANY announced the date; a calendar row or a web date is a projection, served in ``_meta.domain.next_earnings_date_projection_only`` (projected_by lists the sources) with ``_meta.confidence`` `low`. `high` = ≥2 confirming sources or the exchange listing alone, `medium` = one other confirming source. Ties go to the documented source priority, web search last. A results-release date the company announced in its own 8-K is confirmed, `high`, with primary_source issuer_8k, and outranks every calendar; its per_source row carries announcement (form, accession_number, filed, items, document, document_url, filing_url, quote) and event (results_release | earnings_call). An 8-K that dates only the earnings call bounds the release to that day or the business day before: a calendar date there is the answer, and the call date is the answer, `medium`, only when no calendar date is (next_report_resolved.announced_event says which the company dated). A date the company's newest 8-K postponed or cancelled is no source's confirmed date (the issuer_8k row's reason announced_date_withdrawn, with withdrawn). Reason announcement_unclassified: no 8-K dates the report, but one has a sentence about its results dated ahead that the reader could not classify (the row's unclassified: dates, the filing and the sentence); it is not a date, and a date may already be announced. Another company's date in a joint release, and a results release's own call, are not the next report. next_report_resolved.inputs.issuer_8k says whether those filings were read (consulted|unavailable|not_consulted); unavailable = SEC was slow or down, so a date may already be announced; that issuer_8k row carries retryable, retry_at and retry_after_s (when a call reads the filings again). Top-level ``summary`` field for FE rendering. A fund, warrant or preferred line answers ``status: not_applicable`` with ``not_applicable_reason``; a symbol neither the security master nor the market-data listing knows is 404 unknown_ticker; symbol_in_security_master is false (the master lacks it, e.g. a new listing) or null (not checked) with a symbol_check_note. ``next_report_resolved`` is the ONE next-report date shared with intel_street_estimates (status confirmed|projected|unknown|not_applicable; confirmed_by company = the company announced it; source; alternatives; inputs 'not_consulted' for a non-symbol ticker or a not-applicable one). Read the date from next_report_resolved.date. most_recent_reported_date comes from the options-flow calendar whenever it has a report: its last with an actual EPS (most_recent_reported_status calendar_with_actuals), or a newer company-dated report whose results are not in yet (reported_actuals_pending). A date found only in web text is unverified, or unverified_newer_than_calendar when it is newer than the calendar's report and could be the next quarter's (only when the calendar row names its fiscal quarter end). The dates not taken are in most_recent_reported_alternatives, with a calendar date that passed without results (projected_date_passed_no_actuals, company_date_passed_no_results). revenue_estimate is a vendor's consensus line (revenue_estimate_basis), not comparable with filed total revenue. 1h cache.

Latest ML ensemble prediction for one ticker: blended_score, conviction, decile rank, and per-voter sub-scores. Call this when the user asks 'what does the model think of X' or wants a quantitative score to weigh against fundamentals.

UNAVAILABLE: no SHAP drivers can be served. The drivers file does not record which model bundle produced it, so it cannot be tied to the model serving now; every call answers 503 upstream_dead_drivers_unbound until a bound file is published. Top-N SHAP feature attributions for the ML ensemble score on a ticker: drivers[] ranked by |SHAP| with feature (e.g. beta_cma, vol_21d), signed shap_value, direction (bullish/bearish/neutral). PRIMARY tool for 'why is the model bullish/bearish on X?'. Nightly run; default top=5, max 20. 5min cache. Served only when the drivers name the active model bundle and the run and features they were computed on; otherwise a 503 refusal, not billed (error_code upstream_dead_drivers_unbound, `binding.status` unbound | foreign_bundle | mixed_bundles | provenance_incomplete | active_bundle_unknown) — say the drivers are unavailable, never that the ticker is outside the universe, never substitute. Every 200, including available:false, carries as_of, decision_session, model_version, model_bundle_sha256 and `stale` (over 3 days); drivers also carry feature_provenance (model_tier, feature_set_sha256, n_model_features, feature_coverage: the share of the model's features the name was supplied). available:false means only that this run (n_tickers_scored names) holds no drivers for the ticker: not that it is outside the model's universe, and not a statement about its ml_prediction.

Insider transactions straight from the Form 4 filings: owner, title, transaction code, direction, shares, price and shares held after. Call it for insider conviction around events. Transaction codes matter — 'P'/'S' are open-market buys/sells while 'M' is an option exercise and 'A' an award; treating them alike overstates conviction. Coverage is a tracked issuer set, not the whole market: an empty answer says which it is (coverage_status not_covered, for an issuer outside the set, points to intel_insider_flow; no_filings_in_window names the dates held).

Overview

351 read-only tools over one of the deepest institutional financial datasets available to AI agents: US stocks since 1926, over 1 billion implied-volatility surface points since 1996, 21.8M CDS observations and a 10.6M-company private-markets graph. The institutional data desk, inside your AI agent.

Data that usually takes a terminal contract and a stack of vendor subscriptions arrives through one connection, with OAuth sign-in in Claude, ChatGPT, Cursor, VS Code and Codex. Every answer is dated and sourced, and says when a dataset lags.

Credit: 21.8M daily CDS observations since 2005 across 5,517 reference entities, FINRA TRACE corporate bond trades since 2002, credit index composites, 15.8M credit rating actions and 396k syndicated loan facilities with covenants.

Options: over 1 billion vol-surface points since 1996, 92.6M implied borrow rates, 18M dealer greek-exposure records, gamma by strike, unusual flow, and live and replayable chains with greeks.

Fundamentals without look-ahead: 10.2M point-in-time statement versions since 1986, quarterly financials since 1961, about 75 ratios per stock per month since 1970, and every restatement traced.

Estimates and events: 35M analyst-level estimates since 1980, daily consensus since 1978, EPS surprises since 1965, 41.9M corporate events since 1990, and 1.75M transcripts across 21,628 companies.

Ownership and flows: 124M 13F holdings since 1980, 460M mutual fund holdings since 2002, 32M insider transaction records since 1983 split into 10b5-1 and discretionary, 61M securities-lending records since 2002 and short interest since 2006.

Private markets: 355M records covering 10.6M companies, 3M deals, 4.9M people, 645k investors and funds and 22M news items, with fund returns and valuation marks.

Also: 283M scored news events since 2000, about 460 firm characteristics back to 1926, 5M director interlocks, audit and ESG records, 40M geographic revenue records, WARN layoffs since 1989, 68 macro series, CAPE since 1871, futures, FX and 17,000 crypto coins.

Free plan: 1,000 credits a month, no card. Every tool only reads data; nothing places orders or moves money. Data for research, not investment advice.

What you get

351 read-only tools over one of the deepest institutional financial datasets available to AI agents: US stocks since 1926, over 1 billion implied-volatility surface points since 1996, 21.8M CDS observations and a 10.6M-company private-markets graph. The institutional data desk, inside your AI agent.

Data that usually takes a terminal contract and a stack of vendor subscriptions arrives through one connection, with OAuth sign-in in Claude, ChatGPT, Cursor, VS Code and Codex, or a Wealthnow API key in the X-API-Key header. Every answer carries its as-of time, source and coverage, and says when a dataset lags. Every tool is read-only: nothing places orders, moves money or changes your account.

Depth at a glance
US equitiesDaily and monthly stock history since 1926, delistings included; daily bars for about 12,500 tickers since 2003
OptionsOver 1 billion implied-vol surface points since 1996; option volume and open interest since 2000
Credit21.8M daily CDS observations since 2005 (5,517 reference entities); TRACE bond trades since 2002; 15.8M rating actions
Fundamentals10.2M point-in-time statement versions since 1986; quarterly financials since 1961; annual since 1950
Estimates35M analyst-level estimates since 1980; daily consensus since 1978; EPS surprises since 1965
Ownership124M 13F holding records since 1980; 460M mutual fund holding records since 2002; 32M insider records since 1983
Events and text41.9M corporate events since 1990; 1.75M transcripts across 21,628 companies; 283M scored news events since 2000
Private markets355M records: 10.6M companies, 3M deals, 4.9M people, 645k investors and funds, 22M news items
FactorsAbout 460 firm characteristics back to 1926; factor returns since 1926
Macro68 U.S. series as first released; S&P 500 CAPE since 1871

Equities and prices

  • US stock history since 1926: daily and monthly returns, prices, shares outstanding, distributions and delistings, so studies are survivorship-bias-free.
  • Daily bars for about 12,500 US tickers since 2003, with total returns and the full dividend and split history.
  • 325M+ captured intraday minute bars with extended hours, plus 60-second microstructure windows (VWAP, block trades, buy/sell imbalance).
  • A security master joining ticker, CUSIP, ISIN, SEDOL and CIK with dated identifier history.

Credit and fixed income

  • 21.8M daily CDS observations since 2005: 5Y single-name par spreads and implied default probabilities across 5,517 reference entities, through end-2025.
  • FINRA TRACE corporate bond trades since 2002: price, yield, volume and side, matched from the equity ticker.
  • Credit index composites for North America, Europe and Asia.
  • 15.8M credit rating actions, with history for some issuers back to the 1920s.
  • 396k syndicated loan facilities since 1981, with pricing, 2.1M lender shares and 68k financial covenants.

Options and volatility

  • Over 1 billion implied-vol surface points since 1996: standardized surfaces across maturities and deltas for skew and term-structure research.
  • Option volume and open interest since 2000, and 92.6M option-implied borrow rates since 2010.
  • 18M dealer greek-exposure records: gamma in aggregate, by strike and over time, plus market, sector and ETF options tide.
  • Live option chains with full greeks, IV and OI, replayable chains on 5,000+ underlyings, unusual flow, IV rank and VIX since 1990.

Fundamentals without look-ahead

  • 10.2M point-in-time statement versions since 1986: each quarter as first reported, as it stood on any date, with every restatement traced line by line.
  • Quarterly financials since 1961 and annual since 1950, segment data since 1976 and principal customers.
  • About 75 ratios per stock per month since 1970, plus 100+ current metrics and a multi-filter screener.
  • As-filed XBRL history and filings read live from SEC EDGAR, with structured extracts.

Estimates, events and text

  • 35M analyst-level estimates since 1980, broker by broker, plus consensus since 1976, daily consensus since 1978 and EPS surprises since 1965.
  • 7.5M price targets, 3.3M recommendations since 1992 and 2M management guidance records, compared with consensus at the time.
  • Crowd-sourced estimates since 2010 alongside the street.
  • 41.9M corporate events since 1990: M&A, guidance changes, buybacks, executive changes, activism, offerings, index changes and 100+ more types.
  • 1.75M transcripts of 530k events across 21,628 companies since 2002, with full text and 25M speaker turns from 2020.
  • 283M scored news events since 2000, aggregated to daily company sentiment, plus a live newswire.

Ownership, insiders and the short side

  • 124M 13F holding records since 1980, every holder quarter over quarter.
  • 460M mutual fund holding records since 2002 and ETF holdings with inverse lookup.
  • 32M insider transaction records since 1983, split into 10b5-1 plan and discretionary trades, with Form 144 intent-to-sell notices.
  • 61M daily securities-lending records since 2002, short interest since 2006, 52M short-volume records since 2009 and dark-pool prints.
  • Congressional trades, lobbying filings and federal contract awards.

Governance, ESG and alternative data

  • 1.2M board profiles, 871k directors and 5M director interlock links.
  • Executive pay since 1992; audit fees, restatements and auditor changes since 2000; forensic accounting flags.
  • ESG ratings since 2007 with pillar, theme and key-issue scores.
  • 6M+ supply-chain relationships and 40M geographic revenue records since 2003.
  • WARN layoff notices since 1989 and monthly headcount, hiring and attrition.

Private markets

  • 355M records covering 10.6M companies, 3M deals and funding rounds, 4.9M people, 645k investors and funds and 100k+ limited partners.
  • Fund performance (IRR, DPI, TVPI, RVPI), round syndicates, comparables, 22M news items and model valuation marks that refuse to guess when coverage is thin.

Quant research

  • About 460 firm characteristics back to 1926, including the 102 published anomaly characteristics, factor returns since 1926 and rolling factor betas since 1966.
  • 47 research datasets readable directly, and model-ready feature panels with measured coverage and freshness.
  • Deflated Sharpe Ratio and CPCV checks, TWAP and VWAP scenarios, uncertainty-discounted Kelly sizing and a DCF scenario simulator.

Macro, futures, FX and crypto

  • 68 U.S. economic series as first released, S&P 500 CAPE since 1871 and leading indicators for 22 economies.
  • Rates: the Treasury curve across all 11 tenors, auction results, Fed holdings and primary-dealer positioning since 2013.
  • Futures: 28 CME, CBOT, NYMEX and COMEX products, term structures and CFTC Commitments of Traders across 892 contracts since 2006. Ship traffic through 24 maritime chokepoints since 2019.
  • About 17,000 crypto coins with cross-venue funding, open interest, basis, BTC and ETH vol surfaces, DEX pools and on-chain flows. Live FX rates.

Plans and sign-in

  • Free plan: 1,000 credits a month, 30 requests a minute, no card. It covers prices, fundamentals and SEC filings and stops at its grant, so it never bills you. Deep history, credit, private markets, options and transcripts are on paid plans (pricing).
  • Failed calls are free, and account_status and search_tools never spend credits.
  • Finding tools: search_tools finds any of the 351 tools by topic, and call_tool runs any tool your plan includes.

Wealthnow provides data for research and education. It is not investment advice, and Wealthnow is not a broker-dealer or a registered investment adviser.

Connect

Add https://mcp.wealthnow.io/mcp as a remote MCP server in Claude, ChatGPT, Cursor, VS Code or Codex and sign in with OAuth, or send an X-API-Key header. Free plan: 1,000 credits a month, no card.

Comments

More Finance & Commerce MCP servers

Stockpile Market MCP

casualbrooks

Official Model Context Protocol (MCP) server for Stockpile Market - Autonomous Agent Commerce & Curation.

PumpView — Solana Trading & Live Data for AI Agents

PumpView

Solana trading and live trade data for AI agents. Build swaps with flexible slippage, sign locally, submit through PumpView and inspect confirmed receipts. Includes wallet-based API access, per-key volume and USDC data s

Zensei MCP

Zensei

Zensei MCP is a remote MCP server available at https://www.zensei.com.

Fruit Stand Fund Returns

fruitstanddev

Hosted MCP server for trailing and calendar-year total returns of 32,000+ US mutual funds and ETFs — compare, screen, and explain fund performance by ticker.

iDevice Buyer's Guide

idevicewearables

Independent buyer's guide for wearables and phones: buy now or wait, compatibility, prices, release history and news. Every fact is sourced and labelled with how sure we are. Free, read-only, no sign-in.

Advertise