> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pegana.xyz/llms.txt
> Use this file to discover all available pages before exploring further.

# Data model & glossary

> The core asset object, its enums, and the fields you'll actually branch on.

Every read endpoint returns some shape of an **asset** — a tracked mint with its latest
peg reading. This page pins the closed enums and the fields that matter; the full,
authoritative schema is the live [OpenAPI spec](https://api.pegana.xyz/openapi.json).

## Enums (closed sets)

<ResponseField name="state" type="PegState">
  The engine's class-aware verdict — **trust this, don't re-threshold `discount` yourself**.
  `PEGGED` · `DRIFT` · `DEPEG` · `CRITICAL` · `BLACK_SWAN` · `UNKNOWN`. `UNKNOWN` is
  non-alerting (missing/untrustworthy data — never rendered as healthy).
</ResponseField>

<ResponseField name="anchor" type="Anchor">
  How to read fair value: `USD` (fixed \$1 target) · `FX` (fixed non-USD fiat rate, see
  `peg_target`) · `NAV` (intrinsic/redemption value that legitimately moves — LSTs,
  yield-bearing stables, synthetics).
</ResponseField>

<ResponseField name="confidence" type="Confidence">
  Pyth oracle confidence bucket for the reading: `high` · `medium` · `low` · `unknown`.
  Scopes the price-oracle interval only, not market-depth reliability.
</ResponseField>

<ResponseField name="calibration" type="Calibration">
  `calibrated` · `provisional`. The self-describing form of `monitoring_only` — `provisional`
  means the asset is live but its thresholds are still being validated on mainnet data.
</ResponseField>

<ResponseField name="risk_side" type="RiskSide">
  Which direction of `discount` is the risk: `below` (USD/FX anchors — a discount from peg)
  or `either` (NAV anchors — premium and discount both matter).
</ResponseField>

## Key fields

<ResponseField name="mint" type="string">
  Canonical Solana mint — the **stable join key**. Resolve assets by mint, not symbol.
</ResponseField>

<ResponseField name="discount" type="string (decimal)">
  Signed spread vs. fair value, as a fraction (`"0.0123"` = 123 bps). ×10000 → bps. Negative =
  below peg / discount, positive = premium. See [Response format](/guides/rest-api#response-format).
</ResponseField>

<ResponseField name="intrinsic_usd / market_usd" type="string (decimal)">
  The two legs of the discount: mechanism/NAV fair value vs. observed market price.
</ResponseField>

<ResponseField name="risk_score" type="integer | null">
  Derived 0–100 peg-risk score, anchored to `state` (PEGGED ≤ 24 … BLACK\_SWAN 95–100). A single
  sortable number that never disagrees with the classification. `null` when there's no
  trustworthy discount.
</ResponseField>

<ResponseField name="market_cap / circulating_supply" type="string | null">
  Size/context signals (USD market cap, circulating supply). **Never** an input to
  `discount`/`state`/thresholds. `null` until observed.
</ResponseField>

<ResponseField name="worst_abs_24h / jitter_bps_24h" type="string (decimal)">
  Trailing-24h stability: worst absolute discount, and peak-to-trough dispersion (jitter).
  Same signed-fraction units as `discount`.
</ResponseField>

<ResponseField name="monitoring_only" type="boolean">
  `true` = actively monitored but excluded from the calibrated launch set (a calibration
  status, not a health state — a `monitoring_only` asset can still be `PEGGED`).
</ResponseField>

<ResponseField name="updated_at" type="string (RFC3339)">
  Freshness of the current snapshot. The engine stops writing when a feed goes stale, so a
  frozen `updated_at` is how a stale feed surfaces — always gate on it.
</ResponseField>

<Note>
  Values follow the [Response format](/guides/rest-api#response-format) conventions: decimals
  are exact strings (parse with a decimal library), timestamps are RFC3339 millisecond-`Z`.
</Note>