619,831 tools. Updated 2026-09-28 18:53
"XState" matching MCP tools:
- Public mode returns FS AI RMF framework reference data only — not org-specific scoring. Use when assessing an organization FS AI RMF governance maturity stage or preparing a regulatory AI roadmap presentation. Returns INITIAL, MINIMAL, EVOLVING, or EMBEDDED classification with stage criteria and remediation priorities. Example: EVOLVING stage organizations have documented AI policies but lack systematic model validation — typical gap to EMBEDDED is 18-24 months and 12-15 additional controls. Connect org MCP for org-specific scoring. Source: FS AI Risk Management Framework. $0.02 USDC per call.ConnectorNo auth
- Private equity firms (management companies and advisers) as compact cards: the classifier's class with its state, confidence and basis; size band with the evidence it was sized on; fund counts, latest vintage and Form ADV gross assets reported for private equity funds since 2024 (reported gross assets, NOT fund size or dry powder); portfolio and transaction counts (36 and 12 months); team counts; acquisition appetite and investment velocity values. Filter by name, class, class_state, size_band, state or minimum transactions in 36 months. Returns dfx:pe: ids. Funds, people, transactions, scores with their components and counterparties are on get_pe_firm, not on these cards. Not for family offices, independent sponsors or venture firms (search_family_offices, search_independent_sponsors, search_vc_firms). ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- Resolve an advisor by name AND a firm (name or CRD), or by individual CRD, to one person card. A name alone is never resolved to one person: 439,425 registrants share many names, so with only a name the answer lists labelled candidates and `resolved` is false. With a firm name, the current firm is tried first, then prior registrations with that firm. The card that comes back is the same as get_ria_advisor's summary; call get_ria_advisor(dfx_id) for the history. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- Use when a pharmacy, supply chain, or compliance agent needs FDA drug recall status for an NDC or drug name. Returns recall records (status, classification, reason, dates), match_confidence (ndc_exact or name_match), normalized ndc11, and openFDA provenance — never silently returning a different product's recalls. Example: NDC 41163-703-10 — Class II nasal spray recall records with termination dates. Source: openFDA drug/ndc and drug/enforcement. | x402 SLA: $0.10 USDC per call. Returns HTTP 503 (no charge) when upstream data sources unavailable. data_sources[] discloses provenance bound by synthesis.output_hash.ConnectorNo auth
- Lenders ranked on the comparable facilities they CURRENTLY HOLD on BDC schedules: same lien, a borrower industry containing the term (as the filer wrote it: health, software, industrial, business services), a facility size lower bound inside the band, entered since the window. Score = comparables 35%, recency 20%, size fit 15%, mark on the comparable book 15%, sponsors named 15%. Refuses one stale comparable, a book marked under 0.85 and passive holds under $2M. Sizes are lower bounds (BDC pieces), never the commitment. `states` filters on borrower headquarters where DFX holds one (GLEIF, about one comparable in six today); the answer reports geo_coverage, so say how many comparables carried a state. This is the answer to 'find lenders for a $60M unitranche for a sponsor-backed healthcare services business': use it before search_private_credit, whose industry filter lists borrowers. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- Lenders ranked on the commercial real estate loans they ORIGINATED, from CMBS loan-level filings (SEC ABS-EE): same property type, loans in the states and city given, loans inside the size band nationally, recency and size fit. Each row names the lender (originator names normalised to the parent, a loan with several originators credits each), its local and national comparable counts and balances, the latest origination and three example loans with property, city, trust and the SEC filing. This is the answer to 'find lenders for a $100M hotel in Atlanta' or 'who lends on multifamily in Texas': use it, never find_lenders_for_financing, whose comparables are corporate BDC loans. A thin local market is carried by national comparables in the band and the envelope says how many were local. Conduit originators only: balance-sheet bank loans (most construction loans), debt funds and life companies are not on this tape, so pair it with search_bank_cre_exposure for local banks. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
Matching MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables Claude to drive a gated incident-response runbook that enforces step order and logs all actions via an XState machine.Apache 2.0
- AlicenseAqualityAmaintenanceEnables German real-estate appraisal calculations, including property depreciation and purchase-price allocation, through MCP tools backed by AfaMax APIs.2202 npmMIT
Matching MCP Connectors
Free, no key. US commercial real estate: FHA/CMBS maturities, LIHTC/HUD expiries, bank CRE exposure.
7M+ real estate transactions from Poland's RCN registry. Search, compare, and analyze prices.
- Loans on the ACTIVE CMBS tape (SEC ABS-EE, each loan's latest monthly reading, read within 100 days of the newest period) with the servicer's own fields: payment status (current, 30/60/90 days, non performing matured balloon), special servicer and transfer date, whether it is in special servicing now, workout strategy, modification, current DSCR with its basis, occupancy, appraisal and LTV, balance, maturity, originator, and the sponsor as the prospectus annex discloses it (sponsor column, else carve-out guarantor, labelled). Each row carries the core property id where the building chain resolved it, for get_property_record. Filter by state, property type, maturity window, distress or sponsor text, or ask for ONE building with property (its name or street address, e.g. 'Park West Village', '1384 Broadway') or property_dfx_id (the id resolve_address or get_property_record gave). The answer to 'Texas multifamily loans maturing in 18 months that are in special servicing, and who sponsors them'. Conduit CMBS only: bank balance sheet, agency and debt fund loans are not on this tape. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- Every conduit CMBS loan whose servicer reported a foreclosure, REO or deed in lieu workout (SEC ABS-EE), with how it ended: resolved or still pending, the SEC liquidation code decoded (disposition/liquidation, discounted payoff, payoff, repurchase), the liquidation date, the balance before resolution, the last reported realized loss and loss severity, the special servicer, and the trust filing that reported the resolution. Plus a summary by resolution and the total loss reported. The answer to 'which CMBS loans went to foreclosure and how were they resolved'. A zero realized loss is stored as absent, so 'not reported' is never 'no loss'. For live loans and their current status use search_cmbs_loans. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- Use this tool when your workflow needs flatten json for agent memory. Input: value. Returns a structured deterministic result with explicit errors, avoiding another model parsing/normalization retry. Flatten values to JSON Pointer keys. Empty string is root; / is an empty property name. Empty containers stay typed containers, not strings. Limits: request 24 KB, JSON depth 24, 4,000 nodes. Reserved prototype keys and unsafe integers rejected. Inputs processed transiently; private HMAC trace metadata retained.ConnectorNo auth
- Use this tool when your workflow needs deduplicate json records. Input: records, keys. Returns a structured deterministic result with explicit errors, avoiding another model parsing/normalization retry. Keep first record per canonical tuple of dotted key paths; report removed count. All records must contain every key. Null is a value, missing is an error. Limits: request 24 KB, JSON depth 24, 4,000 nodes. Reserved prototype keys and unsafe integers rejected. Inputs processed transiently; private HMAC trace metadata retained.ConnectorNo auth
- Publish a bounded structured artifact for other clients to retrieve by ID or tags. Returns an immutable state ID, content hash and lineage. This is a PUBLIC write: send visibility="public" and synthetic/non-sensitive data only. To derive a new version, retrieve its parent and supply parent_id plus read_receipt in the same application context. Payloads are stored as data and never executed.ConnectorNo auth
- Check a JSON artifact against explicit schema constraints before returning or forwarding it. Returns valid, errors, artifact hash and exact verification scope. A code string or structured plan can be checked for shape/explicit values only: this does not execute code, prove semantics or certify a plan. Optional state_id and read_receipt verify use of an identical previously retrieved public state.ConnectorNo auth
- Family offices as compact cards: class (single, multi, embedded...) with confidence, whether they invest directly, sectors and asset classes on record, check size where stated, direct investment count, latest deployment, and counts of people, sponsor and real estate relationships. A list filter sorted by one column: for a ranked answer to a buyer's criteria use rank_family_offices. Filter by states or region, canonical industries, direct investing, manager allocation, recent activity, AUM, an event type since a date, or a named co-investor. Returns dfx:fo: ids. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- Registered investment advisers (Form ADV: RAUM, clients by type, employees, advisors on IAPD, private funds, class such as INDEPENDENT_WEALTH or WIREHOUSE), registered advisors (IAPD: current firm, registration history counts, exams, designations) and the private funds advisers report, as compact cards with dfx:ria: ids. `query` takes a name, a firm CRD number, an SEC file number (801-...), an individual CRD (with entity_type=advisor) or an SEC private fund id (805-...). Filter firms by state, class, wealth managers only, RAUM band, advisor headcount or private funds; advisors by firm CRD or state; funds by firm CRD or type. Every row names its fact class (reported: filed on Form ADV; fact_from_registration: IAPD registration dates; derived: a rule or arithmetic over filed inputs). RAUM double counts affiliates; no advisor book size exists or is estimated; the IAPD tape is survivor-biased before 2026-09-15. Nothing here is predictive. Not for private equity or venture managers as investors (search_pe_firms, search_vc_firms), though the same adviser may appear on both graphs (get_entity same_as). ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- Use when benchmarking EHR maintenance costs before a contract renewal or evaluating health IT budget efficiency. Returns benchmark cost per licensed bed with optional gap analysis when actual cost and bed count are provided. Example: Epic maintenance median $4,500/bed — 300-bed hospital at $6,200/bed is 38% above market — renegotiation trigger especially strong at 5+ year renewal cycles. Source: KLAS 2024, Kaufman Hall EHR TCO composite. $0.02 USDC per call.ConnectorNo auth
- Forward opportunities across DFX: an entity, the event a measured pattern points to (a borrower reaching non-accrual, a pension plan re-upping with a manager, an RIA entering succession, a nursing home closing), the horizon, the pattern's lift over its base rate with the lower bound of its interval and a band (BASELINE, ELEVATED, HIGHLY_ELEVATED, EXCEPTIONAL, cut on the lower bound), an economic range with its unit, dated why-now facts, the evidence labelled by kind, confidence by dimension, and whether a person is identified. Each row says whether it may be read as 'likely to' (a VALIDATED pattern) or only as an emerging indicator. The likelihood belongs to the pattern, never to the entity: do not restate a lift as the chance this entity acts. Filter by vertical, predicted event, persona, horizon, economic size, band, contact coverage or entity. Rows come ranked by the model's entity rank (rank.position, rank.family_percentile). Use FIRST for any question about what is likely, next, near-term, about to happen, or who shows indicators. Say which status each row has: VALIDATED may be described as a measured pattern; TESTING and PROSPECTIVE_ONLY are emerging indicators and must not be called predictions. An empty answer means no governed forward row matches, not that nothing is developing: then say plainly that DFX has no validated forward signal for it (get_forward_signal_ledger shows what was tested and failed), and never offer completed events (a move that happened, a completed succession) as a forecast. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- The capital parties named on announced independent sponsor transactions, aggregated. group=provider ranks lenders and equity providers by distinct deals and distinct sponsors financed, with how many sponsors they financed more than once, their roles, verticals and first and last deal. group=pair lists sponsor and provider pairs that repeat (min_deals, default 2) with the targets. group=deal returns the deal rows. role_class=debt keeps lenders, senior, mezzanine, unitranche and subordinated debt; equity keeps equity and co-invest. EBITDA filters use the TARGET's estimated EBITDA band (employees and sector ratios, confidence 0.45 to 0.55; no deal states EBITDA) and the answer says how many deals could not be sized. The answer to 'which capital providers are most active in independent sponsor deals', 'who lends to IS deals of $10M to $50M EBITDA' and 'which sponsors and lenders work together repeatedly'. DEALS LIKE ONE DEAL: give deal_sponsor and deal_company ("which lenders financed deals similar to Glen Oaks Capital's acquisition of StanChem Resins") and the deal itself is resolved, then providers on OTHER sponsors' deals in that deal's vertical since `since` (default 730 days ago) are ranked by distinct sponsors backed, each marked if it already backs this sponsor; the same comparable set the capital map letters use. A company that matches none of the sponsor's deals is DEAL_NOT_FOUND with the sponsor's deals listed, never a guessed sector. Returns dfx:isi: ids. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- The observed pre-move indicators DFX holds, per advisor: a prior co-move teammate or the team's senior member left the firm (with who and where to), a fifth or more of the advisor's branch or practice left in six months, a Series 65 or 63 added outside licensing classes, or the advisor owns part of a newly filed adviser. Each indicator is dated, names its source (IAPD registration spans or Form ADV) and carries the back-test that admitted it: the 365-day move rate with and without it, the lift with its 95% interval (only lower bounds of 1.15 or more are served) and the lift inside the move model's deciles. Filter by firm (dfx:ria: id, CRD or name; a name covers every CRD that carries it), advisor, office city, state, kind or date; firm-scoped answers add counts by office and kind. These are indicators, never a forecast that a person will move, and most advisors carrying one do not move within a year. Use this for 'who shows signs of moving', 'who else in that office or team might leave', 'flight risk'; use search_ria_advisor_moves for moves that already happened. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- Resolve a firm, fund, person or company name to canonical dfx ids from the Data Factory's index of every published name and alias (former names, dbas, legal names) on the family office, sponsor, venture, private equity, RIA, allocator, private credit and real estate fund graphs as each graph is indexed. Each row carries the ROLES the institution holds across graphs where the identity layer publishes them, so one institution that is a private equity manager, a registered adviser, a private credit manager, a real estate fund manager and a manager holding allocator capital comes back as one identity with the id on each graph, never as five unrelated firms. Ranked exact, prefix, then word match; one row per entity; typically under 200 ms. The first call before get_entity. Real estate organisations are resolved by resolve_organization instead. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- RIA M&A from three factual sources, each labelled: successions the acquirer swore on Form ADV Item 4 (succession); firms whose advisors re-registered whole at one other firm (absorbed, a rule over registration dates); and firms that left the SEC adviser roster (roster_exit; the reason is not stated). Each row carries the target and, where there is one, the acquirer as dfx:ria: ids, the date, headline, advisor count or RAUM as magnitude, and the source. Filter by kind, acquirer or target CRD, firm id, date window or minimum advisors. Absorbed also returns acquisitions measured from registration history (a succession, half the firm in one 14-day chain, or 25 or more advisors in one month) with the advisors transferred, how many are still registered through the acquirer at 30, 90, 182 and 365 days, and the pooled retention of similar-size deals. No press, no rumours, no predictions. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth
- The RIA event tape by OBSERVATION time: advisor firm changes and departures, team lift-outs and absorptions, successions, RAUM and headcount changes, control persons joining or leaving, first ADV filings, roster exits, each with occurred_at (the registration or filing date), observed_at (when DFX wrote it), fact class, magnitude, previous and new state, evidence and source. `since` is an ISO timestamp; poll with the `next_since` the answer returns. The historical backfill was written on 2026-09-15, so a `since` before that day returns history in bulk; the Data Factory's first-seen accrual for RIA begins when its event arm exists. Filter by event type, signal family (people, ownership, firm, consolidation) or subject id. ACCESS: without a paid DFX plan on the vertical, a list returns its first 5 rows in full and a count of the rest by type (locked.count, locked.by_type), never the rows; a record names its subject and the first 3 related names per section; contact values (email, phone, profile URLs) and decision-maker names are never returned, only their types and counts. Every answer says what it withheld in `entitlement` and `locked`. Full access: DFX Intelligence, 3 days for $1 at https://dfxintel.com/data-factory/plans.ConnectorNo auth