Skip to content
DAMREV
Start a Structuring Review

Guide

Structured Probabilities

Structured probabilities is how the DAMREV platform treats inputs that are not settled facts. External data enters through one engine, values that are views carry human verification, a doubtful reading can stop an action, and decisions that need a person follow set approval paths. Every judgement stays on the programme record, before and after encoding.

The term

What does structured probabilities mean on DAMREV?

It means giving uncertainty a structure instead of hiding it. An instrument depends on some inputs that are readings, some that are views and some that are decisions. The platform names each kind and routes it through the engine or the person responsible for it.

The platform lets each kind hold or stop an action according to how far it can be relied on.

  • Carried by BeaconReadingsData from outside the platform, such as a reserve balance or a market price. A reading can be stale or wrong at its source.
  • Carried by EquilibriumViewsJudgements built on readings, such as a fair market value, with human verification. A view can be reasonable and still be disputed.
  • Carried by approval pathsDecisionsActs by authorised people, such as approving an encoded rule. A decision can be made by the wrong person, or without the information it needed.

The problem

Why is an uncertain input dangerous once it is encoded?

Because encoded rules act without hesitation. A person who doubts a number pauses; a contract executes. If an estimate is encoded as though it were certain, every automatic action that follows inherits the error, and on a network whose ledger is final at close the action cannot be quietly undone afterwards1.

A reserve reading. If treated as certain: Tokens are minted against reserves that are not there. The platform's control: Forge reads Beacon's mint circuit-breaker signal before any issuance.
  • A reserve reading
  • A reserve reading
  • Tokens are minted against reserves that are not there (If treated as certain)
  • Forge reads Beacon's mint circuit-breaker signal before any issuance (The platform's control)
A reserve reading. If treated as certain: Tokens are minted against reserves that are not there. The platform's control: Forge reads Beacon's mint circuit-breaker signal before any issuance.

Structured finance has always depended on uncertain inputs. Tokenization removes many of the steps in which people question them, which is much of its value, but it also removes the pauses in which a doubtful number would have been caught.

Tokenized real-world assets on public chains, excluding stablecoins, were reported at about 26.4 billion US dollars in March 2026, up from about 6.6 billion US dollars a year earlier2.

The answer is not to avoid encoding. Soroban contracts run natively on the same Stellar ledger that settles the asset3. The answer is to decide, before encoding, which inputs a contract may act on alone, which need human verification, and which must be able to stop it.

External data

How does external data reach the platform?

Through Beacon, the proof-of-reserve and oracle engine. Each external reading enters at one point, where it can be examined, and reaches only engines named in advance.

Run it. Oracle network, then Beacon, then Forge, then Anchor, then Equilibrium. Run it to follow one external reading.
  • Run it
  • Oracle network: Designed to bring external data on-chain.
  • Beacon: Receives price and reserve data and supports the mint circuit‑breaker.
  • Forge: Reads the circuit-breaker signal before any mint.
  • Anchor: Prices collateral from Beacon's feeds.
  • Equilibrium: Combines Beacon's data with human verification.
Run it. Oracle network, then Beacon, then Forge, then Anchor, then Equilibrium. Run it to follow one external reading.

Run it to follow one external reading.

Beacon states five consumers: Anchor, Reserve, Registry, Equilibrium and Forge. It depends on no other engine, which keeps the entry point simple to audit, and it makes no price, valuation or return claim about any asset, because a feed is a reading and not an opinion.

Evidence of reserves is also becoming a legal expectation: in the United States, the GENIUS Act of July 2025 requires payment stablecoin issuers to hold reserves on at least a one-to-one basis4.

Human judgement

Where does human judgement enter?

Wherever a value is a view rather than a reading. Equilibrium combines oracle data from Beacon with human verification to express a fair-market-value view and to model token economies, presented as a framework, never as a price, a guarantee or an investment signal.

Equilibrium depends on Beacon and Veritas, so the verification behind a view is tied to a verified person. Seven engines state that they consume Equilibrium's model: Forge, Cadence, Reserve, Flux, Agora, Registry and Portal.

Stops

What happens when an input is in doubt?

The action can stop. Beacon supports a mint circuit-breaker that can halt minting when a reserve or data condition is not met, and Forge reads that signal before any issuance.

  1. Identity clears

    Forge checks identity from Veritas before any mint.

  2. Rules clear

    Sentinel supplies the rule set that governs the asset.

  3. The breaker is read

    Forge reads Beacon's circuit-breaker signal. A failed condition halts the mint.

  4. Issue or stop

    Forge issues under the issuer's authority only when all three have cleared.

Release and transfer carry the same kind of stop. Cadence gates each release through Veritas and Sentinel, and Sentinel returns a decision at every transfer. An input that fails its condition holds the action rather than letting it through.

Approval paths

Who decides when a person must decide?

The people the approval path names. Four eyes and six eyes determine who may assess, approve or review each class of decision, so no one person both does a thing and approves it.

The record

How is every judgement kept traceable?

By attaching it to the programme record. Every decision, input, encoded rule and approval stays with the programme, so the operating instrument can always be traced back to the terms that authorised it.

  • The input an action relied on, and the engine it entered through
  • The view applied to it, and the human verification behind that view
  • The encoded rule, and the attestation on the smart contract
  • The approval, and who gave it under four eyes or six eyes
  • The version of the DAMREV Standard in force at the time

The record answers the question a regulator asks after the event: was this a reasonable action on the information available? Under the January 2026 statement of the Securities and Exchange Commission (SEC), putting a security in token form does not change how the federal securities laws apply to it5.

Worked example

How would a warehoused commodity programme handle a doubtful reserve reading?

An illustrative programme, not a client: tokens represent claims on a commodity held in a warehouse, with reserves reported by an external source.

Illustrative example

A doubtful reserve reading on a warehoused commodity

  1. 01

    Step 1: Before encoding

    The asset is researched under the DAMREV Standard. The team agrees which reserve source Beacon will read, the condition a reading must meet, and which decisions need four eyes or six eyes.
  2. 02

    Step 2: A reading arrives

    Beacon brings the reserve data on-chain through an oracle network. The reading does not meet the agreed condition.
  3. 03

    Step 3: The mint halts

    Beacon's circuit-breaker signal halts minting, and Forge issues nothing further. Transfers of tokens already issued still pass through Sentinel's rules.
  4. 04

    Step 4: People decide

    The discrepancy goes to the approval path agreed for that class of decision. The person who assesses it and the person who approves the response are different people.
  5. 05

    Step 5: The view is revisited

    Where the programme uses a fair-market-value view, Equilibrium's framework is revisited with fresh data and human verification, and presented as a framework, not a price.
  6. 06

    Step 6: The record keeps it

    The reading, the halt, the decision and the approvers stay on the programme record, so a later reviewer can follow what happened and why.

Common questions

What else do people ask?

Short answers on structured probabilities: whether it is a pricing or forecasting model, which engine brings in external data, whether a failed data condition can stop issuance, who verifies a fair-market-value view, whether agentic AI decides on its own, what the DAMREV Standard contributes, and whether any of it is marked as deployed.

Is structured probabilities a pricing or forecasting model?

It is a way of handling inputs, not a model of prices. The platform makes no price, return or yield statement. Equilibrium expresses fair market value as a framework, and Beacon supplies data without any valuation claim.

Which engine brings external data onto the platform?

Beacon, through an oracle network. The engines that need external price or reserve data read it from Beacon: Anchor, Reserve, Registry, Equilibrium and Forge.

Can a failed data condition stop issuance?

Yes. Beacon supports a mint circuit-breaker that can halt minting when a reserve or data condition is not met, and Forge reads the signal before any issuance.

Who verifies a fair-market-value view?

People do, alongside oracle data. Equilibrium combines Beacon's data with human verification, and it depends on Veritas, so the verification is tied to a verified person.

Does agentic AI make decisions on its own?

Agents act inside defined roles. Praxis runs agents that use the engines as tools, routed through Sentinel and Veritas, signed via Sentry, and composed through the roles held in Stratum. Four eyes and six eyes apply to agents as they do to people.

What does the DAMREV Standard contribute?

It sets how an asset is researched, how human judgement is applied before encoding begins, and how assessment is separated from approval. Each version is dated, and superseded versions stay readable.

Is any of this marked as deployed?

Capability status is recorded separately, engine by engine, and those records are being prepared. This page describes how the platform is designed to treat uncertain inputs and marks no capability as live.

Where next

The pages below pick up where this one ends. Start with the one closest to where your asset or programme stands today.

References

  1. 1.Chainstack, "What is Stellar". https://chainstack.com/what-is-stellar/ Also: Stellar developer documentation, "Control access to an asset", https://developers.stellar.org/docs/tokens/control-asset-access
  2. 2.PYMNTS, "Tokenized real-world asset value jumps fourfold to $26 billion", reporting RWA.xyz data. https://www.pymnts.com/blockchain/2026/tokenized-real-world-asset-value-jumps-fourfold-to-26-billion/ Also: RWA.xyz, live data, https://app.rwa.xyz/
  3. 3.Stellar Development Foundation, "Protocol 20 and smart contracts are live on Mainnet". https://stellar.org/blog/developers/protocol-20-and-smart-contracts-are-live-on-mainnet
  4. 4.Greenberg Traurig, "GENIUS Act enacted, establishing a regulatory framework for payment stablecoins". https://www.gtlaw.com/en/insights/2025/7/genius-act-enacted-establishing-a-regulatory-framework-for-payment-stablecoins-issued-or-sold-in-the-united-states
  5. 5.US Securities and Exchange Commission, Statement on Tokenized Securities, 28 January 2026. https://www.sec.gov/newsroom/speeches-statements/corp-fin-statement-tokenized-securities-012826-statement-tokenized-securities

Last updated:

Ready to build on programmable infrastructure?

Tell us what you want to tokenize. We will show you how DAMREV can run it.