TokenBank — tokenized real-world assets

Find TokenBank answer pages for a natural-language question — e.g. "can I buy Tesla as a token", "best stablecoin yield on Base", "euro stablecoins MiCA", "pre-IPO OpenAI token", "fixed yield on-chain". Returns the best-matching questions with the first sentence of each computed answer (figures from the live dataset), the page URL to cite and a JSON URL with every row, plus the matching guides (explainers such as how a bot buys a token, how to verify a contract, what happens if an issuer fails), each with a JSON URL. A question in Arabic, German, Spanish, Japanese, Chinese, Korean, Thai, Vietnamese or Indonesian also returns our pages written in that language (localPages), e.g. "صكوك مرمّزة". Use this first for a question a person would type; use the other tools to filter beyond it.

Search and filter TokenBank's instruments for earning yield with crypto. Searches the CURATED tier by default (the rows that state a yield, minimum and risk); pass tier to reach the imported issuer-catalogue rows (tokenized treasuries, yield stablecoins, gold, real estate, tokenized stocks, staking, DeFi/CeFi earn, alternative funds). Returns compact records; use get_product_details for the full record.

Full record for one instrument by ticker (case-insensitive): yield, risk, custody, availability, how to buy, source link, verification date.

Start here when building or running an AI agent or trading bot that buys tokens on its own. Returns the whole how-to as data: the six steps from wallet to confirmed purchase with the MCP/API call for each, the purchase paths (wallet only / owner onboarded once / platform account) with live counts, which asset classes a self-custody wallet can reach without an identity check, the highest-yield wallet-only candidates, how to choose among them and when not to buy (howToDecide: checks naming the fields that answer them, not recommendations), the guardrails a wallet policy should enforce, and the guides that explain it (each with a JSON URL). Then call get_purchase_steps for one ticker.

Use this before buying more than one product: spreads an amount over several sources of yield instead of putting everything into the single highest yield. Returns the blocks (one product per category, never the same exposure twice, each checked at its own amount incl. gas, weight capped), a 10 % admixture above the risk budget for extra yield, the median yield and risk per category, the blended yield vs. the single best, the worst single failure as % of the amount, and a guide written for agents: why spreading is better (with failures on record), the steps (then get_purchase_steps per block) and when to rebalance. Checks, not advice.

How an AI agent or bot buys one instrument on its own, step by step: the purchase path (wallet only / owner onboarded once / platform account), the three KYC gates, network and verified contract, the issuer route, the measured trading venues, the exit terms where researched (KYC to redeem, minimum, fee, settlement, from issuer documentation), the ordered steps (fund, verify contract, approve exact amount, simulate, deposit or swap, confirm, exit) and the guardrails a wallet policy should enforce. Read-only: it does not build, sign or send a transaction, and it does not say who may buy.

Compare low-risk, instant-access crypto savings instruments against a typical bank savings account rate. Returns the honest trade-offs, not just the bigger number.

Instruments whose secondary market needs no KYC and which a self-custody wallet holds itself — a property of how the token works, read from hand-verified gates. For real-world assets the gate to MINT and the gate to trade on a secondary market routinely differ, so a fund can be permissioned at the issuer while its token trades freely. Strict by design — an instrument whose gate is merely unverified is excluded. This says nothing about who may buy it: the issuer's terms and local law decide that.

Put two or more instruments side by side on the fields that decide between them: yield, risk, minimum, custody, the three KYC gates, target/area, issuer and verification date. Use for "X vs Y" questions instead of calling get_product_details repeatedly.

Shape of the dataset: how many instruments, split by pillar, category and risk, how many need no KYC on the secondary market, and how fresh the figures are. Use to ground a claim about coverage before making one.

Takes no parameters.

Which countries this dataset can see into, which it cannot, and why — the question "what is tokenized in Japan / Korea / Brazil / India" answers from here rather than from a general search. This matters because the honest answer is uncomfortable: roughly $9bn of tokenized securities sits in markets NO Western aggregator indexes, this one included. Japan holds over ¥1tn on a permissioned consortium chain run by eighteen banks. Korea issued $2.2bn in 2026 under regulatory sandbox exemptions, with legal recognition phased in from 4 February 2027 and retail capped by amount PER VENUE. India, Vietnam, Indonesia and Thailand have no stablecoin in their own currency at all. Call this before stating what a market holds, or before quoting any total as if it were the world. Omit `country` for every market.

Official registers of tokenized assets and digital-asset firms for any of 250 jurisdictions (ISO 3166 + EU): which authority supervises, which public registers exist, which we read, and the entries themselves. Use it for "what is registered in Lithuania", "is firm X licensed anywhere" (pass `query` without `country`), or "which countries publish a list of token issues" (kind=issuance). A register entry records a licence, registration or approval by an authority — it is NOT proof that a token was issued, and it says nothing about who may buy. Provider registers list companies, not instruments: never add their entries to a count of tokens.

Drill into the RealT portfolio: the individual tokenized US rental properties behind the REALT entry, each with its own net rental yield (computed as annual net rent / tokenized property value), token price, monthly rent and occupancy. Use this when someone asks which specific tokenized property to buy, or for the yield/price spread within one issuer.

Does a tokenized version of a given company, ETF or bond exist, and from which issuer? Searches the whole imported tier by company name, ticker or ISIN: issuer catalogues from Backed / xStocks, Ondo Global Markets, Dinari, Robinhood, Coinbase, Backpack, Reality (Bitget), ST0x, Jarsy, Midas, Lofty, Swiss SME share registers from Aktionariat, Japanese security tokens, El Salvador's CNAD register (registered offerings — sale not confirmed, say so) and European and Asian digital bonds from official registers, plus an aggregated tier (CoinGecko) covering the rest. Use this BEFORE saying a stock is not available on-chain. THREE THINGS MATTER IN THE ANSWER: (1) many companies are issued SEVERAL times over — Apple exists as seven different tokens — and those are not duplicates. They differ in issuer, legal form, holder rights, chain and the market each issuer aims them at (targetArea). Report the issuer, not just that the token exists. (2) Sources differ in AUTHORITY. Issuer catalogues state what that issuer issued. The aggregated rows only observe that a token exists: they carry no KYC gate and no custody claim. targetArea is a description of the issuer's stated audience, never an eligibility check. (3) This is identity data: the source states no yield, no minimum and no risk grade, so those fields are absent rather than zero. For instruments that do carry a comparable rate, use search_yield_products instead.

For a security tokenized by more than one issuer, compare every token on it side by side — issuer, structure, KYC gates, custody, chain, and the target/area each issuer states — plus every Hyperliquid perpetual on it (fees, funding, leverage; no ownership). The right tool for "which tokenized versions of Apple exist?" and "where can I get NVIDIA on-chain, and at what cost?": the tokens and perps share a price driver and differ in almost everything else. Securities that exist on-chain only as a perp are included too. Hundreds of securities here are tokenized more than once — Apple exists as seven separate tokens; the response carries the exact counts, do not quote a figure from this description. targetArea describes who each issuer aims the product at and where; it is not an eligibility check — who may buy is set by the issuer. Omit `query` to list every such security.