Redemption Backstop Changelog: v2 archive
Historical release evidence. Return to current versions and the complete history index.
What this controls
Redemption Backstop Changelog: v2 archive 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
Semantics correction for non-deterministic HOLLAR exit
A route-semantics review removes one overstated redemption path and explicitly leaves several harder assets outside medium-confidence coverage until a credible holder backstop is established.
Impact snapshot
- HOLLAR is no longer modeled as a `psm-swap` redemption route because the Hydration Stability Module only guarantees buying HOLLAR from the facility, while protocol buybacks of HOLLAR remain opportunistic rather than holder-deterministic
- The harder follow-up set led to no new medium-confidence additions for crvUSD, sUSD, MIM, or USDU Finance because current public materials still do not establish a primary redemption rail comparable to the existing modeled route families
- This keeps redemption coverage honest by preferring uncovered or low-coverage states over overstated direct-exit semantics
v2.9Mar 24, 2026Semantics correction for non-deterministic HOLLAR exit
A route-semantics review removes one overstated redemption path and explicitly leaves several harder assets outside medium-confidence coverage until a credible holder backstop is established.
- HOLLAR is no longer modeled as a `psm-swap` redemption route because the Hydration Stability Module only guarantees buying HOLLAR from the facility, while protocol buybacks of HOLLAR remain opportunistic rather than holder-deterministic
- The harder follow-up set led to no new medium-confidence additions for crvUSD, sUSD, MIM, or USDU Finance because current public materials still do not establish a primary redemption rail comparable to the existing modeled route families
- This keeps redemption coverage honest by preferring uncovered or low-coverage states over overstated direct-exit semantics
DetailsHide details
Semantics correction for non-deterministic HOLLAR exit
A route-semantics review removes one overstated redemption path and explicitly leaves several harder assets outside medium-confidence coverage until a credible holder backstop is established.
- HOLLAR is no longer modeled as a `psm-swap` redemption route because the Hydration Stability Module only guarantees buying HOLLAR from the facility, while protocol buybacks of HOLLAR remain opportunistic rather than holder-deterministic
- The harder follow-up set led to no new medium-confidence additions for crvUSD, sUSD, MIM, or USDU Finance because current public materials still do not establish a primary redemption rail comparable to the existing modeled route families
- This keeps redemption coverage honest by preferring uncovered or low-coverage states over overstated direct-exit semantics
Impact notes
- HOLLAR is no longer modeled as a `psm-swap` redemption route because the Hydration Stability Module only guarantees buying HOLLAR from the facility, while protocol buybacks of HOLLAR remain opportunistic rather than holder-deterministic
- The harder follow-up set led to no new medium-confidence additions for crvUSD, sUSD, MIM, or USDU Finance because current public materials still do not establish a primary redemption rail comparable to the existing modeled route families
- This keeps redemption coverage honest by preferring uncovered or low-coverage states over overstated direct-exit semantics
v2.8Mar 24, 2026Second medium-confidence redemption cleanup tranche
A second cleanup tranche upgrades the best non-top-100 low-confidence routes where Pharos already had sufficient issuer, reserve, or queue-redemption evidence to stop relying on heuristics.
- cUSD, cEUR, ALUSD, and AZND now use reviewed eventual or reserve-backed redemption semantics instead of heuristic capacity ratios
- USDA now carries a reviewed issuer-redemption route, while pUSD Plume now uses live reserve metadata with a documented 1:1 USDC fallback instead of a generic low-confidence issuer assumption
- Names whose route semantics are still genuinely unresolved, such as crvUSD, sUSD, MIM, and HOLLAR, remain outside this tranche rather than being promoted on weak evidence
DetailsHide details
Second medium-confidence redemption cleanup tranche
A second cleanup tranche upgrades the best non-top-100 low-confidence routes where Pharos already had sufficient issuer, reserve, or queue-redemption evidence to stop relying on heuristics.
- cUSD, cEUR, ALUSD, and AZND now use reviewed eventual or reserve-backed redemption semantics instead of heuristic capacity ratios
- USDA now carries a reviewed issuer-redemption route, while pUSD Plume now uses live reserve metadata with a documented 1:1 USDC fallback instead of a generic low-confidence issuer assumption
- Names whose route semantics are still genuinely unresolved, such as crvUSD, sUSD, MIM, and HOLLAR, remain outside this tranche rather than being promoted on weak evidence
Impact notes
- cUSD, cEUR, ALUSD, and AZND now use reviewed eventual or reserve-backed redemption semantics instead of heuristic capacity ratios
- USDA now carries a reviewed issuer-redemption route, while pUSD Plume now uses live reserve metadata with a documented 1:1 USDC fallback instead of a generic low-confidence issuer assumption
- Names whose route semantics are still genuinely unresolved, such as crvUSD, sUSD, MIM, and HOLLAR, remain outside this tranche rather than being promoted on weak evidence
v2.7Mar 24, 2026Buffer-backed medium-confidence redemption tranche
A follow-up tranche promotes the remaining cleanest heuristic routes by tying their capacity bounds to already-curated stable redemption buffers or direct full-reserve rails.
- USDD, LISUSD, reUSD, USR, USDF, DUSD, USP, and BUCK now use reviewed documented-bound capacity instead of generic heuristic ratios because Pharos already tracks explicit stable redemption buffers for those routes
- msUSD and fxUSD now carry reviewed direct-redemption semantics rather than unresolved low-confidence defaults, reflecting Main Street's full USDC reserve rail and f(x)'s documented collateral redemption path
- Routes whose reserve stack still lacks a clearly bounded redeemable stable buffer, such as YUSD, USN, and UTY, intentionally remain low-confidence until the evidence improves
DetailsHide details
Buffer-backed medium-confidence redemption tranche
A follow-up tranche promotes the remaining cleanest heuristic routes by tying their capacity bounds to already-curated stable redemption buffers or direct full-reserve rails.
- USDD, LISUSD, reUSD, USR, USDF, DUSD, USP, and BUCK now use reviewed documented-bound capacity instead of generic heuristic ratios because Pharos already tracks explicit stable redemption buffers for those routes
- msUSD and fxUSD now carry reviewed direct-redemption semantics rather than unresolved low-confidence defaults, reflecting Main Street's full USDC reserve rail and f(x)'s documented collateral redemption path
- Routes whose reserve stack still lacks a clearly bounded redeemable stable buffer, such as YUSD, USN, and UTY, intentionally remain low-confidence until the evidence improves
Impact notes
- USDD, LISUSD, reUSD, USR, USDF, DUSD, USP, and BUCK now use reviewed documented-bound capacity instead of generic heuristic ratios because Pharos already tracks explicit stable redemption buffers for those routes
- msUSD and fxUSD now carry reviewed direct-redemption semantics rather than unresolved low-confidence defaults, reflecting Main Street's full USDC reserve rail and f(x)'s documented collateral redemption path
- Routes whose reserve stack still lacks a clearly bounded redeemable stable buffer, such as YUSD, USN, and UTY, intentionally remain low-confidence until the evidence improves
v2.6Mar 23, 2026Moderate-effort redemption confidence tranche
A moderate-effort tranche reviews a final group of already-modeled lower-confidence routes where Pharos now has stronger primary redemption semantics, but not yet protocol-native live instant-buffer telemetry across the full set.
- DOLA and JupUSD now treat their published stable-buffer bounds as reviewed documented-capacity inputs instead of leaving those ratios in the heuristic bucket
- rwaUSDi, mTBILL, MUSD, USDN, and YZUSD now use reviewed redemption semantics with documented-bound capacity instead of generic low-confidence placeholders, while YUSD, USN, and UTY keep conservative bounded-capacity assumptions because their delta-neutral collateral stacks still lack explicit published live buffers
- The documented-bound subset now contributes medium-confidence redemption evidence, while the reviewed delta-neutral routes stay visible-only until Pharos has explicit buffer bounds or live telemetry
DetailsHide details
Moderate-effort redemption confidence tranche
A moderate-effort tranche reviews a final group of already-modeled lower-confidence routes where Pharos now has stronger primary redemption semantics, but not yet protocol-native live instant-buffer telemetry across the full set.
- DOLA and JupUSD now treat their published stable-buffer bounds as reviewed documented-capacity inputs instead of leaving those ratios in the heuristic bucket
- rwaUSDi, mTBILL, MUSD, USDN, and YZUSD now use reviewed redemption semantics with documented-bound capacity instead of generic low-confidence placeholders, while YUSD, USN, and UTY keep conservative bounded-capacity assumptions because their delta-neutral collateral stacks still lack explicit published live buffers
- The documented-bound subset now contributes medium-confidence redemption evidence, while the reviewed delta-neutral routes stay visible-only until Pharos has explicit buffer bounds or live telemetry
Impact notes
- DOLA and JupUSD now treat their published stable-buffer bounds as reviewed documented-capacity inputs instead of leaving those ratios in the heuristic bucket
- rwaUSDi, mTBILL, MUSD, USDN, and YZUSD now use reviewed redemption semantics with documented-bound capacity instead of generic low-confidence placeholders, while YUSD, USN, and UTY keep conservative bounded-capacity assumptions because their delta-neutral collateral stacks still lack explicit published live buffers
- The documented-bound subset now contributes medium-confidence redemption evidence, while the reviewed delta-neutral routes stay visible-only until Pharos has explicit buffer bounds or live telemetry
v2.5Mar 23, 2026Reviewed docs-backed quick-win redemption tranche
A docs-backed quick-win tranche upgrades nine existing low-confidence redemption routes where the remaining blocker was heuristic capacity or stale access and fee assumptions rather than missing telemetry.
- avUSD, cUSD, USDu, cgUSD, HONEY, EUSD, AID, OUSD, and USBD now use reviewed documented-bound redemption capacity instead of staying low-confidence under heuristic supply models
- USDu and AID now reflect whitelist-gated direct redemption access, while cgUSD and AID also disclose reviewed live fee assumptions from official docs
- These routes still do not claim a separately measured live instant buffer, but they now contribute medium-confidence redemption evidence across roughly half a billion dollars of additional tracked market cap
DetailsHide details
Reviewed docs-backed quick-win redemption tranche
A docs-backed quick-win tranche upgrades nine existing low-confidence redemption routes where the remaining blocker was heuristic capacity or stale access and fee assumptions rather than missing telemetry.
- avUSD, cUSD, USDu, cgUSD, HONEY, EUSD, AID, OUSD, and USBD now use reviewed documented-bound redemption capacity instead of staying low-confidence under heuristic supply models
- USDu and AID now reflect whitelist-gated direct redemption access, while cgUSD and AID also disclose reviewed live fee assumptions from official docs
- These routes still do not claim a separately measured live instant buffer, but they now contribute medium-confidence redemption evidence across roughly half a billion dollars of additional tracked market cap
Impact notes
- avUSD, cUSD, USDu, cgUSD, HONEY, EUSD, AID, OUSD, and USBD now use reviewed documented-bound redemption capacity instead of staying low-confidence under heuristic supply models
- USDu and AID now reflect whitelist-gated direct redemption access, while cgUSD and AID also disclose reviewed live fee assumptions from official docs
- These routes still do not claim a separately measured live instant buffer, but they now contribute medium-confidence redemption evidence across roughly half a billion dollars of additional tracked market cap
v2.4Mar 23, 2026Maple syrup withdrawal route correction
Maple's syrupUSDC and syrupUSDT routes now model the documented withdrawal queue instead of an overstated near-instant redemption buffer.
- syrupUSDC and syrupUSDT now use reviewed queue-redemption semantics with documented-bound eventual capacity rather than a heuristic 30% immediate buffer assumption
- Access is now modeled as whitelisted onchain, reflecting Maple's PoolPermissionManager gating for `requestRedeem` and `redeem` calls
- These routes now contribute medium-confidence redemption evidence while preserving Maple's documented FIFO processing and potential multi-day settlement delay
DetailsHide details
Maple syrup withdrawal route correction
Maple's syrupUSDC and syrupUSDT routes now model the documented withdrawal queue instead of an overstated near-instant redemption buffer.
- syrupUSDC and syrupUSDT now use reviewed queue-redemption semantics with documented-bound eventual capacity rather than a heuristic 30% immediate buffer assumption
- Access is now modeled as whitelisted onchain, reflecting Maple's PoolPermissionManager gating for `requestRedeem` and `redeem` calls
- These routes now contribute medium-confidence redemption evidence while preserving Maple's documented FIFO processing and potential multi-day settlement delay
Impact notes
- syrupUSDC and syrupUSDT now use reviewed queue-redemption semantics with documented-bound eventual capacity rather than a heuristic 30% immediate buffer assumption
- Access is now modeled as whitelisted onchain, reflecting Maple's PoolPermissionManager gating for `requestRedeem` and `redeem` calls
- These routes now contribute medium-confidence redemption evidence while preserving Maple's documented FIFO processing and potential multi-day settlement delay
v2.3Mar 23, 2026Reviewed lower-cap redemption cleanup tranche
A small lower-cap cleanup tranche upgrades Pleasing and Apyx routes from generic heuristics to reviewed redemption semantics without adding new live telemetry assumptions.
- PUSD and PGOLD now use reviewed documented-bound redemption routes tied to Pleasing's published off-ramp and physical-delivery docs instead of generic heuristic issuer assumptions
- apxUSD now reflects the documented whitelist-gated mint/redeem rail rather than a generic permissionless stablecoin-redeem assumption
- These routes still do not claim a separately measured live instant buffer, but they now contribute medium-confidence redemption evidence instead of remaining low-confidence heuristics
DetailsHide details
Reviewed lower-cap redemption cleanup tranche
A small lower-cap cleanup tranche upgrades Pleasing and Apyx routes from generic heuristics to reviewed redemption semantics without adding new live telemetry assumptions.
- PUSD and PGOLD now use reviewed documented-bound redemption routes tied to Pleasing's published off-ramp and physical-delivery docs instead of generic heuristic issuer assumptions
- apxUSD now reflects the documented whitelist-gated mint/redeem rail rather than a generic permissionless stablecoin-redeem assumption
- These routes still do not claim a separately measured live instant buffer, but they now contribute medium-confidence redemption evidence instead of remaining low-confidence heuristics
Impact notes
- PUSD and PGOLD now use reviewed documented-bound redemption routes tied to Pleasing's published off-ramp and physical-delivery docs instead of generic heuristic issuer assumptions
- apxUSD now reflects the documented whitelist-gated mint/redeem rail rather than a generic permissionless stablecoin-redeem assumption
- These routes still do not claim a separately measured live instant buffer, but they now contribute medium-confidence redemption evidence instead of remaining low-confidence heuristics
v2.2Mar 23, 2026Live-buffer routes for Ethena and Falcon synthetics
USDe and USDf now reuse live reserve telemetry for current redeemable stable buffers, turning two large synthetic-dollar gaps into reviewed route coverage.
- USDe now models the whitelisted direct mint-and-redeem rail documented by Ethena, with fresh live Liquid Cash telemetry used as the current redeemable stable buffer and a conservative 0.5% fallback bound when telemetry is unavailable
- USDf now models Falcon's KYC-only queued redemption route with a live stablecoin-buffer input from Falcon's transparency feed and a reviewed zero protocol-fee assumption based on Falcon docs
- These routes materially expand medium-confidence redemption coverage without pretending either protocol has a permanently fixed instant-exit buffer
DetailsHide details
Live-buffer routes for Ethena and Falcon synthetics
USDe and USDf now reuse live reserve telemetry for current redeemable stable buffers, turning two large synthetic-dollar gaps into reviewed route coverage.
- USDe now models the whitelisted direct mint-and-redeem rail documented by Ethena, with fresh live Liquid Cash telemetry used as the current redeemable stable buffer and a conservative 0.5% fallback bound when telemetry is unavailable
- USDf now models Falcon's KYC-only queued redemption route with a live stablecoin-buffer input from Falcon's transparency feed and a reviewed zero protocol-fee assumption based on Falcon docs
- These routes materially expand medium-confidence redemption coverage without pretending either protocol has a permanently fixed instant-exit buffer
Impact notes
- USDe now models the whitelisted direct mint-and-redeem rail documented by Ethena, with fresh live Liquid Cash telemetry used as the current redeemable stable buffer and a conservative 0.5% fallback bound when telemetry is unavailable
- USDf now models Falcon's KYC-only queued redemption route with a live stablecoin-buffer input from Falcon's transparency feed and a reviewed zero protocol-fee assumption based on Falcon docs
- These routes materially expand medium-confidence redemption coverage without pretending either protocol has a permanently fixed instant-exit buffer
v2.1Mar 23, 2026Mid-cap route correction and review tranche
A mid-cap tranche adds missing USX, USDa, and M redemption configs while correcting USD.AI and NUSD onto reviewed routes that better match their protocol docs.
- USX, USDa, and M now carry reviewed redemption routes instead of remaining uncovered, and NUSD now uses reviewed documented-bound queue semantics rather than a generic 20% heuristic
- USD.AI now models the base token's direct burn-and-withdraw stablecoin rail instead of inheriting the slower sUSDai unstaking assumptions
- These routes still do not claim a separately measured live instant buffer, but they materially expand medium-confidence redemption coverage across the mid-cap queue
DetailsHide details
Mid-cap route correction and review tranche
A mid-cap tranche adds missing USX, USDa, and M redemption configs while correcting USD.AI and NUSD onto reviewed routes that better match their protocol docs.
- USX, USDa, and M now carry reviewed redemption routes instead of remaining uncovered, and NUSD now uses reviewed documented-bound queue semantics rather than a generic 20% heuristic
- USD.AI now models the base token's direct burn-and-withdraw stablecoin rail instead of inheriting the slower sUSDai unstaking assumptions
- These routes still do not claim a separately measured live instant buffer, but they materially expand medium-confidence redemption coverage across the mid-cap queue
Impact notes
- USX, USDa, and M now carry reviewed redemption routes instead of remaining uncovered, and NUSD now uses reviewed documented-bound queue semantics rather than a generic 20% heuristic
- USD.AI now models the base token's direct burn-and-withdraw stablecoin rail instead of inheriting the slower sUSDai unstaking assumptions
- These routes still do not claim a separately measured live instant buffer, but they materially expand medium-confidence redemption coverage across the mid-cap queue
v2.0Mar 23, 2026Third lower-cap redemption review tranche
A third lower-cap review tranche upgrades more issuer-style routes from heuristic supply-full modeling to reviewed documented-bound capacity and corrects frxUSD onto its direct onchain stablecoin redemption rail.
- thBILL, XAUm, USDGO, and USA₮ now carry reviewed documented-bound eventual redemption capacity instead of generic heuristic supply-full modeling
- XAUm now discloses a reviewed 25 bps redemption fee and T+3 settlement expectations, while USDGO now uses the reviewed zero-fee StableHub exchange rail documented by OSL
- frxUSD now models the direct onchain USDC mint/redeem contract path as a reviewed stablecoin-redeem route instead of sitting in a generic offchain issuer bucket
DetailsHide details
Third lower-cap redemption review tranche
A third lower-cap review tranche upgrades more issuer-style routes from heuristic supply-full modeling to reviewed documented-bound capacity and corrects frxUSD onto its direct onchain stablecoin redemption rail.
- thBILL, XAUm, USDGO, and USA₮ now carry reviewed documented-bound eventual redemption capacity instead of generic heuristic supply-full modeling
- XAUm now discloses a reviewed 25 bps redemption fee and T+3 settlement expectations, while USDGO now uses the reviewed zero-fee StableHub exchange rail documented by OSL
- frxUSD now models the direct onchain USDC mint/redeem contract path as a reviewed stablecoin-redeem route instead of sitting in a generic offchain issuer bucket
Impact notes
- thBILL, XAUm, USDGO, and USA₮ now carry reviewed documented-bound eventual redemption capacity instead of generic heuristic supply-full modeling
- XAUm now discloses a reviewed 25 bps redemption fee and T+3 settlement expectations, while USDGO now uses the reviewed zero-fee StableHub exchange rail documented by OSL
- frxUSD now models the direct onchain USDC mint/redeem contract path as a reviewed stablecoin-redeem route instead of sitting in a generic offchain issuer bucket