rwa_multichain.token_metadata

rwa_multichain.token_metadata describes the real-world product behind each token in rwa_multichain.tokens: what backs it, who issues it, which platform brought it onchain, and the legal and regulatory structure holders actually buy into. Grain: one row per token per chain, keyed on (blockchain, token_id). It covers roughly 2,300 products from 218 issuers across 125 tokenization platforms.

Why it is separate from `tokens`

tokens answers “does this token exist and what is its identifier on each chain”. token_metadata answers “what is the product, and how is it structured”. Keeping them apart means the registry can add a chain deployment without waiting on research, and research can be revised without rewriting the registry. A product and a token are not the same thing. One product — say Franklin Templeton’s BENJI fund — has a row per chain it is deployed on, and the product-level columns repeat identically across those rows. Deduplicate on product_id when you are counting products rather than deployments.

Table schema

Token identity

Column	Type	Description
`blockchain`	`VARCHAR`	Chain the token instance lives on
`token_id`	`VARCHAR`	Normalized token identifier. Join key to `tokens`, `balances`, `transfers`, and `supply`
`token_standard`	`VARCHAR`	Chain-specific standard: `erc20`, `spl`, `stellar_asset`, `xrpl_iou`, `aptos_fa`, `sui_coin`
`symbol`	`VARCHAR`	The token’s own onchain symbol, as returned by `symbol()` or the chain equivalent

Product identity

Column	Type	Description
`product_id`	`VARCHAR`	Stable lowercase product identifier, unique per product. Often resembles the token symbol but is not resolved from it
`product_name`	`VARCHAR`	The product’s official name as used by its issuer or platform
`issuer_id`	`VARCHAR`	Stable identifier for the legal issuing entity, shared across every product it issues
`issuer_name`	`VARCHAR`	Canonical legal issuer name, written identically everywhere it appears
`platform_id`	`VARCHAR`	Stable identifier for the tokenization or distribution platform
`platform_name`	`VARCHAR`	Canonical platform display name

Asset classification

Column	Type	Description
`asset_class`	`VARCHAR`	Top-level category of the backing exposure. See values below
`asset_type`	`VARCHAR`	Finer classification, constrained by `asset_class`. See values below
`underlying_asset`	`VARCHAR`	Description of specifically what backs the token, precise enough to verify against a primary source
`underlying_ticker`	`VARCHAR`	Market ticker of the underlying instrument, where one trades independently of the token. Not the token’s own symbol
`underlying_exchange`	`VARCHAR`	Listing venue for `underlying_ticker`, with its ISO 10383 MIC where known — e.g. `Nasdaq (XNAS)`. Disambiguates ticker collisions across venues

Product mechanics and lifecycle

Column	Type	Description
`product_design`	`VARCHAR`	Plain-language narrative of how the product is issued, held, moved, and redeemed
`transfer_restriction`	`VARCHAR`	What the token contract enforces on transfers: `NONE`, `ALLOWLIST`, `PLATFORM_INTERNAL`, `NON_TRANSFERABLE`
`acquisition_access`	`VARCHAR`	How a fresh wallet can come to hold it: `PERMISSIONLESS`, `KYC_ONBOARDING`, `PLATFORM_ACCOUNT_ONLY`
`transfer_model`	`VARCHAR`	Free-text detail behind the two columns above — allowlisting rules, cross-chain burn-and-mint behavior, DEX eligibility
`redemption_mechanism`	`VARCHAR`	How a holder gets liquidity out: `INSTANT_ONCHAIN`, `T_PLUS_N_OFFCHAIN`, `SECONDARY_MARKET_ONLY`
`product_status`	`VARCHAR`	Lifecycle state: `ACTIVE`, `MATURED`, `WOUND_DOWN`, `DEFAULTED`

Jurisdiction and legal structure

Column	Type	Description
`jurisdiction_country`	`VARCHAR`	Domicile of the issuing legal entity — not the platform’s, manager’s, or custodian’s
`jurisdiction_iso2`	`VARCHAR`	Two-letter ISO country code matching `jurisdiction_country`
`legal_wrapper_type`	`VARCHAR`	Legal form through which holders obtain exposure. See values below
`asset_manager_entity`	`VARCHAR`	Entity managing the underlying portfolio, when distinct from the issuer — e.g. Amundi for Spiko funds
`custodian_entity`	`VARCHAR`	Institution holding the underlying non-digital asset: a bank or trust company for securities and cash, a vault operator for physical commodities
`regulatory_framework`	`VARCHAR`	The specific exemption, registration, or regime the offering relies on
`regulator`	`VARCHAR`	Supervising authority as a canonical short name. Shared acronyms are disambiguated by country — `FMA (Liechtenstein)` vs `FMA (Austria)`
`issuer_entity`	`VARCHAR`	Exact legal registrant named in this product’s filing. Can be more specific than `issuer_name`, e.g. a named sub-fund
`issuer_lei`	`VARCHAR`	The issuer’s 20-character GLEIF Legal Entity Identifier
`issuer_cik`	`VARCHAR`	The issuer’s numeric SEC filer identifier, where applicable
`eligible_investors`	`VARCHAR`	Investor class legally permitted to hold the product. See values below

Research provenance

Column	Type	Description
`confidence`	`VARCHAR`	`HIGH` (primary evidence, no material gaps), `MEDIUM` (primary source found, some detail unconfirmed), `LOW` (inference or secondary evidence remains material)
`verification_status`	`VARCHAR`	`VERIFIED` when every material populated claim has been checked, otherwise `PARTIAL`
`flagged_for_human_review`	`VARCHAR`	Blank when nothing is outstanding; otherwise a short statement of the open question. Treat a populated row as unsettled
`notes`	`VARCHAR`	Caveats, aliases, and dated correction history. Corrections are appended rather than overwriting, so the field carries an audit trail
`researched_date`	`DATE`	Date the row’s research was performed. Not a source-freshness timestamp
`sources`	`ARRAY(ROW)`	Ordered evidence records, capped at ten per product. See below

Blank versus NOT_APPLICABLE

Two sentinels carry different meanings and you should filter on them differently:

Blank means the value is not currently known for this product.
NOT_APPLICABLE means the field has been shown not to apply — a basket product has no single underlying_ticker, a non-SEC issuer has no issuer_cik.

No other placeholder ever appears: there is no N/A, UNKNOWN, or UNDISCLOSED in this table. So WHERE custodian_entity <> '' gives you products with a known custodian, and WHERE custodian_entity = 'NOT_APPLICABLE' gives you products that genuinely have none. asset_class and asset_type are populated for effectively every product. The deep legal fields are populated where a primary source confirms them, so expect blanks on custodian_entity, issuer_lei, and issuer_cik for long-tail products.

asset_class values

Value	Meaning
`credit`	Private credit, loans, receivables, or debt exposure
`equities`	Public or private company shares, or equity funds
`commodities`	Physical or commodity-linked goods
`cash_equivalent`	T-bills, government debt, and money-market instruments
`real_estate`	Property or property-backed interests
`other`	Exposure that does not fit the preceding classes

By product count, credit and equities dominate — private credit facilities and tokenized stock lineups both run to hundreds of products each — while cash_equivalent holds far fewer products but most of the AUM.

asset_type values

asset_type is constrained by its parent class:

Parent class	Values
`credit`	`corporate_bonds_high_yield`, `corporate_bonds_investment_grade`, `credit_fund`, `private_credit`, `private_credit_fund`, `private_credit_senior_secured_loans`, `securitized_products`
`equities`	`equities_cash`, `equities_exchange_traded_funds`, `equities_fund`, `equities_index_fund`, `equities_preferred`, `equities_synthetic`, `multi_asset_allocation_fund`, `private_equity`, `venture_capital_fund_interest`
`commodities`	`agriculture`, `carbon_credits`, `gemstones`, `metals`
`cash_equivalent`	`government_debt_non_us`, `government_debt_us`, `money_market_funds_non_us`, `money_market_funds_us`
`real_estate`	`residential_real_estate`
`other`	`crypto_hedge_fund`, `fine_art`, `hedge_fund`, `insurance_linked_securities`

Two values are deliberately cross-class, assigned from the product’s actual exposure rather than inferred from the value alone: energy appears under both commodities and credit, and fixed_income_diversified under cash_equivalent, credit, and other. Filter on the pair when this matters, not on asset_type alone.

legal_wrapper_type values

Value	Meaning
`DIRECT_TOKENIZED_EQUITY`	The token itself is the legal equity interest
`LEDGER_BASED_SECURITY`	The DLT-native register is the legal register of title
`SPECIAL_PURPOSE_VEHICLE`	A product-specific SPV holds the asset
`REGISTERED_FUND`	A formally registered or regulated collective fund
`ETF_PASSTHROUGH_WRAPPER`	The token wraps shares of an existing ETF
`STRUCTURED_NOTE`	A debt instrument contractually tracking an underlying
`STRUCTURED_CREDIT_TRANCHE`	A securitized tranche backed by pooled credit
`DERIVATIVE_CONTRACT`	Synthetic exposure without ownership of the underlying
`CUSTODIAL_RECEIPT`	A receipt representing a specifically custodied asset
`WRAPPED_RWA_TOKEN`	A wrapper around another tokenized RWA product
`DEFI_VAULT_SHARE`	A share of a smart-contract vault or pool
`OTHER`	A structure outside the taxonomy, explained in `product_design`

This is one axis only. It is distinct from regulatory_framework (which regime the offering relies on) and jurisdiction_country (where the issuer is domiciled) — a BVI-domiciled fund can be a REGISTERED_FUND under SEC oversight.

eligible_investors values

UNRESTRICTED, US_RETAIL, US_ACCREDITED, US_QUALIFIED_PURCHASER, NON_US_REG_S, EU_PROFESSIONAL_QUALIFIED, KYC_GATED_GENERAL. Two products from the same issuer can differ here, because eligibility follows each offering’s own terms rather than the issuer. A direct government-bills fund distributed broadly under Reg S and that same issuer’s swap-based fund restricted to EU professional investors will not share a value.

The sources array

sources holds the evidence behind the row, ordered and capped at ten records per product. Each record has these fields:

Field	Type	Description
`source_order`	`INTEGER`	Stable order, contiguous from 1
`url`	`VARCHAR`	Exact page the claim was read from
`label`	`VARCHAR`	Short human-readable citation label
`authority`	`VARCHAR`	`PRIMARY`, `PRIMARY_LEGAL`, `REGISTRY`, `EXPLORER`, or `SECONDARY`
`verifies`	`ARRAY(VARCHAR)`	Column names this source specifically supports
`status`	`VARCHAR`	Retrieval state: `ACTIVE`, `BLOCKED`, `REMOVED`, `ERROR`, or `MISSING`
`last_verified_at`	`DATE`	Date the URL and its claim scope were last checked
`content_hash`	`VARCHAR`	SHA-256 of the normalized scraped page, for change detection

Because verifies is scoped per column, you can check whether a specific field on a specific product is backed by a primary legal source rather than trusting the row-level confidence alone.

Issuer, platform, and asset manager

Three separate entities, and they diverge often enough that picking the wrong one skews a league table:

issuer_name is the legal entity that issues the product. Attribute AUM here for asset-manager rankings.
platform_name is the venue that brought it onchain. Attribute here when measuring tokenization-infrastructure market share. Ondo Global Markets is the platform for tokenized iShares ETFs whose issuer is BlackRock.
asset_manager_entity runs the underlying portfolio when that is a third party — Amundi manages the Spiko funds, Franklin Advisers manages BENJI.

Do not join on symbol

Token symbols are not a safe product key. The same product carries different tickers across access channels, unrelated products share tickers, and scam tokens squat real tickers on other chains. token_id resolves each onchain instance to exactly one product, so join on (blockchain, token_id) and never on symbol or underlying_ticker.

Example queries

AUM, holders, and product count by asset class:

SELECT
  m.asset_class,
  COUNT(DISTINCT m.product_id) AS products,
  COUNT(DISTINCT b.address) AS holders,
  SUM(b.balance_usd) AS aum_usd
FROM rwa_multichain.balances b
JOIN rwa_multichain.token_metadata m
  ON m.blockchain = b.blockchain
  AND m.token_id = b.token_id
WHERE b.day = CURRENT_DATE - INTERVAL '1' DAY
  AND b.balance > 0
GROUP BY 1
ORDER BY 4 DESC NULLS LAST

Issuer league table for tokenized money market funds:

SELECT
  m.issuer_name,
  m.platform_name,
  m.custodian_entity,
  COUNT(DISTINCT m.product_id) AS products,
  SUM(b.balance_usd) AS aum_usd
FROM rwa_multichain.balances b
JOIN rwa_multichain.token_metadata m
  ON m.blockchain = b.blockchain
  AND m.token_id = b.token_id
WHERE b.day = CURRENT_DATE - INTERVAL '1' DAY
  AND b.balance > 0
  AND m.asset_type IN ('money_market_funds_us', 'money_market_funds_non_us')
GROUP BY 1, 2, 3
ORDER BY 5 DESC NULLS LAST
LIMIT 25

Products a US retail investor can hold, by legal wrapper and regulator:

SELECT
  m.legal_wrapper_type,
  m.regulator,
  m.jurisdiction_iso2,
  COUNT(DISTINCT m.product_id) AS products
FROM rwa_multichain.token_metadata m
WHERE m.eligible_investors = 'US_RETAIL'
  AND m.legal_wrapper_type <> ''
GROUP BY 1, 2, 3
ORDER BY 4 DESC

Only well-evidenced rows, for a research or ratings model:

SELECT
  product_id,
  product_name,
  issuer_name,
  asset_class,
  asset_type,
  legal_wrapper_type,
  custodian_entity,
  redemption_mechanism,
  researched_date
FROM rwa_multichain.token_metadata
WHERE verification_status = 'VERIFIED'
  AND confidence = 'HIGH'
  AND flagged_for_human_review = ''
  AND blockchain = 'ethereum'
ORDER BY researched_date DESC

Curated Data

Solana

Hyperliquid

Non-EVM Networks

Protocols

EVM Networks

Partner Data

Dune Index

rwa_multichain.token_metadata

Why it is separate from `tokens`

Table schema

Blank versus NOT_APPLICABLE

asset_class values

asset_type values

legal_wrapper_type values

eligible_investors values

The sources array

Issuer, platform, and asset manager

Do not join on symbol

Example queries

​Why it is separate from tokens

​Table schema

​Blank versus NOT_APPLICABLE

​asset_class values

​asset_type values

​legal_wrapper_type values

​eligible_investors values

​The sources array

​Issuer, platform, and asset manager

​Do not join on symbol

​Example queries

Why it is separate from `tokens`

Table schema

Blank versus NOT_APPLICABLE

asset_class values

asset_type values

legal_wrapper_type values

eligible_investors values

The sources array

Issuer, platform, and asset manager

Do not join on symbol

Example queries