Stability Index Changelog
Full version history of PSI methodology decisions, from v1.0 to v3.67.
What this controls
Stability Index Changelog version history
Use this changelog to trace how Pharos changed the related scoring, data-source, threshold, or interpretation rules over time. Entries record the contract active at each release; archived rules are retained so historical charts and citations can be read against the rules active at the time.
Primary references
Latest version
Canonical historical repair and unavailable-input retention
PSI v3.67 can change repaired historical scores by using one canonical replay path, paired supply identities and the post-audit eligible event universe. Missing required trend or DEWS history preserves accepted days rather than fabricating flat growth or zero stress; formula weights and condition bands are unchanged.
Impact snapshot
- Audit-triggered repairs and bounded backfills share daily prices, native-domain evidence, DEWS archives and persisted input provenance. Event exclusion or restoration recomputes PSI from the post-audit eligible universe in the same atomic transaction.
- Historical seven-day trend compares only paired admitted supply identities, independently of the current market-cap denominator. Observed zero remains valid, but an unavailable or unusable prior denominator holds the accepted day instead of inventing flat growth.
- Missing required v3 DEWS archives remain unavailable and preserve accepted days; an observed all-CALM archive still contributes measured zero stress. Archive date and row count remain part of replay provenance.
- Every historical entrypoint loads the same full twenty-one-day supply lookback, combining the seven-day trend interval and fourteen-day as-of margin, so narrow replay windows do not truncate the eligible cohort.
v3.67Oct 10, 2026Canonical historical repair and unavailable-input retention
PSI v3.67 can change repaired historical scores by using one canonical replay path, paired supply identities and the post-audit eligible event universe. Missing required trend or DEWS history preserves accepted days rather than fabricating flat growth or zero stress; formula weights and condition bands are unchanged.
- Audit-triggered repairs and bounded backfills share daily prices, native-domain evidence, DEWS archives and persisted input provenance. Event exclusion or restoration recomputes PSI from the post-audit eligible universe in the same atomic transaction.
- Historical seven-day trend compares only paired admitted supply identities, independently of the current market-cap denominator. Observed zero remains valid, but an unavailable or unusable prior denominator holds the accepted day instead of inventing flat growth.
- Missing required v3 DEWS archives remain unavailable and preserve accepted days; an observed all-CALM archive still contributes measured zero stress. Archive date and row count remain part of replay provenance.
- Every historical entrypoint loads the same full twenty-one-day supply lookback, combining the seven-day trend interval and fourteen-day as-of margin, so narrow replay windows do not truncate the eligible cohort.
DetailsHide details
Canonical historical repair and unavailable-input retention
PSI v3.67 can change repaired historical scores by using one canonical replay path, paired supply identities and the post-audit eligible event universe. Missing required trend or DEWS history preserves accepted days rather than fabricating flat growth or zero stress; formula weights and condition bands are unchanged.
- Audit-triggered repairs and bounded backfills share daily prices, native-domain evidence, DEWS archives and persisted input provenance. Event exclusion or restoration recomputes PSI from the post-audit eligible universe in the same atomic transaction.
- Historical seven-day trend compares only paired admitted supply identities, independently of the current market-cap denominator. Observed zero remains valid, but an unavailable or unusable prior denominator holds the accepted day instead of inventing flat growth.
- Missing required v3 DEWS archives remain unavailable and preserve accepted days; an observed all-CALM archive still contributes measured zero stress. Archive date and row count remain part of replay provenance.
- Every historical entrypoint loads the same full twenty-one-day supply lookback, combining the seven-day trend interval and fourteen-day as-of margin, so narrow replay windows do not truncate the eligible cohort.
Impact notes
- Audit-triggered repairs and bounded backfills share daily prices, native-domain evidence, DEWS archives and persisted input provenance. Event exclusion or restoration recomputes PSI from the post-audit eligible universe in the same atomic transaction.
- Historical seven-day trend compares only paired admitted supply identities, independently of the current market-cap denominator. Observed zero remains valid, but an unavailable or unusable prior denominator holds the accepted day instead of inventing flat growth.
- Missing required v3 DEWS archives remain unavailable and preserve accepted days; an observed all-CALM archive still contributes measured zero stress. Archive date and row count remain part of replay provenance.
- Every historical entrypoint loads the same full twenty-one-day supply lookback, combining the seven-day trend interval and fourteen-day as-of margin, so narrow replay windows do not truncate the eligible cohort.
- Contributor attribution preserves the full-precision scoring factor without rewriting legacy factors. Displayed thirty-day extrema and averages use source-anchored UTC calendar coverage rather than row count; neither changes the scoring formula.
- Historical repairs are score-moving. Fixed-clock replay, attribution of changed days and owner review are release prerequisites; this entry does not claim a completed replay or rewrite of stored history.
- Activation is provisionally 2026-10-11 00:00 UTC (1791676800); re-date effectiveAt at release
- eb605e5ca
- 718253076
- 52e5df7b2
v3.66Oct 7, 2026Observed supply admission and explicit partial-input quality
Live PSI excludes unavailable current supply from market-cap denominators and contributors, pairs only usable historical supply, and reports successful partial publication as ok with input-quality findings.
- Absent, empty, wholly invalid, nonfinite or negative current totals are omitted; explicit observed zero remains admissible
- Unavailable previous-week totals are omitted from both trend sides; a nonpositive or nonfinite paired denominator holds the prior sample with trend-inputs-unavailable instead of inventing zero trend
- Stored snapshots and the API disclose supplyUnavailableIds and trendUnavailableIds alongside existing unpriced-open-event diagnostics
- A successful sample and prune returns psi-sample-published with ok plus quality for omissions; failed dependencies still hold publication
DetailsHide details
Observed supply admission and explicit partial-input quality
Live PSI excludes unavailable current supply from market-cap denominators and contributors, pairs only usable historical supply, and reports successful partial publication as ok with input-quality findings.
- Absent, empty, wholly invalid, nonfinite or negative current totals are omitted; explicit observed zero remains admissible
- Unavailable previous-week totals are omitted from both trend sides; a nonpositive or nonfinite paired denominator holds the prior sample with trend-inputs-unavailable instead of inventing zero trend
- Stored snapshots and the API disclose supplyUnavailableIds and trendUnavailableIds alongside existing unpriced-open-event diagnostics
- A successful sample and prune returns psi-sample-published with ok plus quality for omissions; failed dependencies still hold publication
Impact notes
- Absent, empty, wholly invalid, nonfinite or negative current totals are omitted; explicit observed zero remains admissible
- Unavailable previous-week totals are omitted from both trend sides; a nonpositive or nonfinite paired denominator holds the prior sample with trend-inputs-unavailable instead of inventing zero trend
- Stored snapshots and the API disclose supplyUnavailableIds and trendUnavailableIds alongside existing unpriced-open-event diagnostics
- A successful sample and prune returns psi-sample-published with ok plus quality for omissions; failed dependencies still hold publication
- Six-hour USD/native price-evidence and 3600-second accepted-DEWS budgets, pure formula and historical rows are unchanged
Earlier versions
Compact summaries retain historical citation targets. Open a version archive for its full evidence and commit provenance.
- v3.65 · Persisted current native-peg event quotes
Oct 6, 2026
Live PSI reads directly observed native quotes retained by the existing depeg producer, without adding network requests to the DB-only lane.
- v3.64 · Null daily components without fabricated zero
Sep 28, 2026
Daily PSI snapshots persist missing components as null while retaining all-day score averages and per-component observation counts.
- v3.63 · Quote-domain-safe event inputs and daily provenance
Sep 27, 2026
Live and historical PSI preserve native event quote domains; daily all-sample averages disclose methodology and component sample counts.
- v3.62 · As-of historical supply lookup
Sep 22, 2026
The historical replay now reads the most recent supply snapshot at or before the day being rebuilt instead of the nearest snapshot in either direction.
- v3.61 · Paired 7-day trend inputs
Aug 22, 2026
PSI now excludes assets without observed previous-week supply from both sides of the 7-day market-cap trend ratio instead of treating them as zero.
- v3.6 · Core-market aggregate universe
Jul 15, 2026
PSI now measures the core stablecoin market without counting parent-linked variants or investment products as independent monetary supply.
- v3.5 · Exact DEWS generation dependency
Jun 19, 2026
PSI now reads the exact DEWS generation advertised by the published-generation pointer instead of selecting latest rows at or before that timestamp.
- v3.4 · Fail-closed DEWS freshness dependency
Jun 6, 2026
PSI now requires a non-empty, fresh latest DEWS row set before publishing a new sample, preventing stale or wiped stress signals from being treated as zero stress breadth.
- v3.3 · Active-only PSI universe
May 14, 2026
PSI eligibility now resolves to active tracked stablecoins plus PSI-only shadow assets, excluding pre-launch and frozen tracked rows from live PSI denominators and repair scopes.
- v3.2 · Fail-closed depeg input availability
Mar 23, 2026
PSI no longer publishes a fresh sample when the active-depeg input query is unavailable, preventing false continuity from an incomplete core dependency.
- v3.1 · Open-depeg replay-price fallback
Mar 23, 2026
Active depegs now stay in PSI when the current stablecoins snapshot temporarily loses a usable price but a recent replay-safe price-cache entry still exists.
- v3.0 · DEWS stress breadth component
Mar 1, 2026
Added DEWS-derived stress breadth as an explicit PSI component to capture broad non-depeg stress.
- v2.1 · Trend hardening and retention safety
Feb 27, 2026
Hardened trend input handling and operationalized rolling retention for PSI samples.
- v2.0 · Removed freezes, rebalanced caps
Feb 26, 2026
Major methodology revision removing freezes and reallocating penalty capacity.
- v1.3 · Sample architecture and 24h average model
Feb 26, 2026
PSI moved to sample-first storage with daily snapshots and 24h average surfaced as the primary displayed signal.
- v1.2 · 15-minute chained compute + depeg depreciation/dedup
Feb 25, 2026
Operational and methodological upgrade to live 15-minute PSI with chronic-depeg depreciation and per-coin deduplication.
- v1.1 · Current deviation semantics
Feb 25, 2026
Severity began using live current deviation rather than event peak deviation for active depegs.
- v1.0 · Initial PSI release
Feb 25, 2026
Launched PSI compute, API, cron persistence, and frontend integration.