> For the complete documentation index, see [llms.txt](https://maxim-speransky.gitbook.io/doc/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://maxim-speransky.gitbook.io/doc/gitbook-risk-framework/gitbook-asset-scoring.md).

# Asset Scoring

Asset Scoring evaluates whether a collateral asset is sufficiently sound and observable to enter the market-selection process. It assesses the asset's economic claim independently of the lending market in which the asset may later be used.

This public page describes the decision architecture. Detailed scoring anchors, validation cases, and operating thresholds remain part of the internal methodology.

## Outputs

```
Eligibility: Eligible / Excluded / Unassessable
Asset Score: 0–100 if Eligible; otherwise N/A
Score band: derived from Allocation Policy if a score exists
Confidence: High / Medium / Low if a score exists
```

* **Eligibility** determines whether the asset may proceed to scoring and downstream policy.
* **Asset Score** provides an ordinal comparison of eligible assets.
* **Confidence** expresses the strength of the evidence supporting the assessment.
* **Score band** is assigned by [Allocation Policy](/doc/gitbook-risk-framework/gitbook-allocation-policy.md) rather than by this scorecard.

`Excluded` means that an explicit knockout condition is present. `Unassessable` means that a critical claim or dependency cannot be established from available evidence. Neither result receives a numeric score, and neither can be overridden by favorable market configuration.

## What is assessed

The unit of analysis is the complete system required for the holder's economic claim to remain valid:

```
Asset → underlying claim → contracts and controls
      → custody and redemption → critical dependencies
```

The analysis therefore extends beyond the token contract. Depending on the asset, it may include an underlying asset, wrapper, bridge, staking system, yield strategy, maturity mechanism, custodian, issuer, servicer, or legal claim.

Each asset receives one primary claim type and every applicable dependency modifier. Primary types cover native network assets, protocol or economic tokens, fiat-backed claims, crypto-backed or synthetic claims, and tokenized RWA claims. Modifiers identify additional layers such as wrapping, bridging, staking, yield generation, principal-token structures, and off-chain dependencies.

This compositional approach prevents a derivative asset from inheriting only the strengths of its underlying. Each affected metric takes the weakest result across the primary claim and applicable dependency lenses. All seven top-level metrics remain active; asset type changes the evidence lens, not the weights. For example, a principal token backed by an RWA yield product must be assessed as a maturity claim, a yield-bearing structure, and an off-chain asset system—not as a standalone token.

## Eligibility before scoring

Eligibility is tested before any weighted calculation. The assessment stops when a critical failure or evidence gap is found.

Typical outcomes include:

* an unidentified critical contract, underlying asset, or dependency results in `Unassessable`;
* unsupported or contradictory evidence about ownership, backing, solvency, or redemption results in `Unassessable`;
* an unresolved exploit, insolvency, material backing mismatch, or failed required redemption path results in `Excluded`;
* authority to create unbounded supply or remove backing without an effective constraint results in `Excluded`.

A freeze, pause, upgrade, or blacklist power is not automatically disqualifying. Its effect depends on whether the power can invalidate the economic claim and whether credible constraints or remedies exist. The internal methodology defines the complete type- and dependency-specific tests.

## Scoring model

Only an `Eligible` asset receives an Asset Score.

| Area                       | Weight | Assessment focus                                                                           | Aggregation                |
| -------------------------- | -----: | ------------------------------------------------------------------------------------------ | -------------------------- |
| **Issuer & Control Risk**  |    40% | Technical authority, governance concentration, accountability, and enforceable constraints | Weighted assessment        |
| **Credit & Backing Risk**  |    35% | Solvency and claim integrity under relevant stress scenarios                               | Scenario-led assessment    |
| **Operational Resilience** |    25% | Implementation assurance, operating history, and economic observability                    | Weakest critical component |

```
Asset Score = Issuer & Control Risk × 0.40
            + Credit & Backing Risk × 0.35
            + Operational Resilience × 0.25
```

### Issuer & Control Risk

This area asks what critical actions are technically possible, who can authorize them, and what transparent or enforceable constraints protect the holder. The relevant lens changes with the claim: a native asset is assessed through monetary and consensus control, while an issuer-backed asset also requires analysis of accountable entities and the legal structure supporting the claim.

### Credit & Backing Risk

This area tests whether the economic claim remains valid under adverse conditions. Relevant scenarios may include reserve impairment, dilution, collateral drawdown, liquidation or hedge failure, governance capture, counterparty default, legal enforcement failure, redemption stress, and reflexive loss of confidence. Dependencies that share the same collateral, operator, governance, or liquidity source are treated as correlated.

The result follows the weakest credible scenario outcome after existing protections; it is not an average of unrelated checks.

### Operational Resilience

This area assesses whether the current system is supported by implementation assurance, meaningful operating history, and evidence that keeps its economic state observable. A strong result in one component cannot average away a critical weakness in another.

The weights are policy parameters based on expert judgment. The score is not a probability of default, an expected-loss estimate, or a substitute for asset-specific analysis.

## Evidence and confidence

Every conclusion must be tied to evidence capable of proving the claim being assessed. On-chain state and deployed code support technical claims; governing documents and official records support legal claims; independent reports and reconciliations support reserve or liability claims. A dashboard or issuer statement may assist discovery but does not automatically establish the underlying claim.

The assessment records the asset identity, dependencies, as-of date, evidence, rationale, and material conflicts. Evidence must correspond to the current deployment and must be refreshed at a cadence appropriate to the speed at which the risk can change.

Confidence is reported separately from the score:

* **High** — critical claims are supported by current primary evidence and meaningful independent corroboration;
* **Medium** — critical claims are established, but some evidence has limited independence, freshness, or completeness;
* **Low** — the claim can still be bounded, but the supporting evidence has material limitations.

Missing or contradictory evidence on a critical claim produces `Unassessable`, not Low confidence. Confidence cannot increase the score. Overall Confidence is the weakest Confidence among critical metrics and dependencies. Economic Observability measures whether the asset exposes the state needed to detect loss; Confidence measures the evidence used in this assessment. The same weakness is not mechanically penalized twice without explaining these distinct effects.

## Assessment flow

```
Classify → map dependencies → test eligibility
         → score eligible assets → record confidence → pass downstream
```

The published assessment shows the result, the material drivers, and the evidence basis. It does not reproduce the complete analyst checklist or the internal decision rules used at each step.

## Policy boundary

Asset Scoring does not determine:

* whether a particular lending market is usable;
* the market oracle, LLTV, interest-rate model, or liquidation route;
* strategy eligibility or portfolio allocation;
* exposure, liquidity, or concentration caps;
* Subordination Class created by overcollateralization.

These decisions belong to [Market Scoring](/doc/gitbook-risk-framework/gitbook-market-scoring.md) and [Allocation Policy](/doc/gitbook-risk-framework/gitbook-allocation-policy.md). Subordination Class can describe additional position-level protection but never changes the intrinsic Asset Score. Shared Morpho protocol architecture is treated as a framework scope assumption rather than re-scored for every asset.

## Model limitations

* The score is ordinal and depends on analyst judgment; it is not a loss forecast.
* Public evidence can be incomplete, delayed, or supplied by an interested party.
* Historical survival does not validate newly introduced code, governance, custody, or backing.
* The assessment can become stale after a material change to the asset or any critical dependency.