Talk to usBook Demo

Index Services

Public preliminary period · not an official benchmark

Index governance

Governance for indices administered by Philidor Labs. This page summarizes the registered methodology; the content-hashed methodology document is the authority wherever the two could disagree.

Index committee

The index committee consists of the two founders of Philidor Labs. The external seat is deliberately open: it will be filled when the right person is found, not for appearance. Until an external methodology review exists, the compensating controls are the methodology published in full, dual-engine reconciliation of every close, signed evidence packages, and the limitations stated honestly in the methodology itself.

Conflicts of interest

Philidor operates no product tracking PHIL10, and none is planned. Risk-assessment commercial relationships with constituents’ operators, where they exist, are disclosed as a category. The controlling rule admits no exception:

“money can buy coverage, custom products and calculation services; it can never buy a risk score or a seat in a standard index”PHIL10 methodology, §14 Governance — quoted verbatim

Methodology changes

Methodology changes are new versions with new content hashes — a registered version is immutable, because compositions and observations reference the version that governed them. Silently rewriting one would retroactively change the rules a signed series was produced under.

  • Every version’s SHA-256 hash is taken over the exact bytes of the document — the published document and the registered hash are the same artefact by construction. See the methodology library.
  • Non-emergency changes carry a public consultation period before they take effect, announced by notice. That obligation is live: it was conditional on the index being public, and the index is public as of methodology v1.5.0. An emergency change may take effect immediately and is announced by notice with its justification.

Notices

  1. statusPublic preliminary period opensphil10-status-preliminary-1.5.02026-08-10

    The registry status of `phil10` moved from `draft` to `preliminary`. Governing methodology: 1.5.0 Content hash: c258381be622c6079695fe0f58580200b5dd66df68428c4c45f153064875a27c Opened by: joel This opens the PUBLIC PRELIMINARY PERIOD described in §11.1 of that document. It is not official inception: no base date is set, no rebase factor is applied, no observation's publication state changed, and no observation is presented as `official`. No benchmark claim is made.

  2. reconstitutionClassification note: 2026-08-04 reconstitutionreconstitution-2026-08-04-classificationunpublished

    The reconstitution effective 2026-08-04 was an operator-invoked run executed immediately after the registration of methodology v1.4.0 (content hash 3d40cb4e7a91da0b29e03949836a4876957b3e693ad47d750f7736261bc6bd88), which added the Aave v3 adapter class and its withdrawability conditions. The recorded reason 'calendar' is imprecise: 2026-08-04 is not a scheduled quarter-end date and no §7.2 extraordinary trigger fired. The run applied the SCHEDULED-run semantics deliberately — full gate set and the §5.1 incumbency buffer — and the resulting membership change (aave-1-0x23878914efe38d27c4d67ab83ed1b93a74d4086a in; morpho-8453-0xee8f4ec5672f09119b96ab6fb59c27e1b7e44b61 out) cleared the buffer on census evidence. A methodology_change reconstitution reason is being added to the codebase so future adapter-class or parameter-change reconstitutions are labeled precisely. The composition, census, and this notice are all insert-only records.

  3. correctionPPR fixings completed after back-calculationppr-backcalc-refixunpublished

    Fixings for live dates whose 30-day window start predated inception were insufficient_data by construction. Back-calculated history (methodology §8.5) supplied those endpoints; each affected date gains a next-revision fixing computed from the exact prior observations, with the endpoint history classes disclosed in the basis. No fixing value was altered — only insufficient_data rows are superseded.

  4. correctionEvidence packages regenerated at schema v2evidence-schema-v2-regenerationunpublished

    The signed evidence packages for closes accepted before schema v2 carried only close outputs. Each affected date gains a next-revision package assembled from the persisted acceptance rows (sealed manifest, reconciled engine pair, composition, methodology), marked regeneratedFrom. Original artifacts remain unchanged.

Corrections

The observation store is insert-only. An error is corrected by new revisions applied as a correction batch, plus a numbered public notice stating what changed, why, which value dates are affected, and what the levels were before and after. Prior revisions remain queryable forever, and the methodology version and composition that governed each revision remain attached to it.

Honest gaps are permanent: once a later close has been accepted, a skipped date is never retroactively filled — doing so would break the internal chaining of an already-signed series.

Cessation

If an index stops being calculated, that is announced with a notice, and the historical series remains published and verifiable. Cessation removes the future, not the record.

Important notices

  1. Public preliminary period. These pages publish a preliminary series. PHIL10 is not an official benchmark and has no official inception date. Official inception is a separate, later act requiring a base date, a completed public preliminary period, and the governance apparatus in the methodology actually operating. Being readable here is not publication in that sense: no observation on this surface isofficial, and none will be until inception.
  2. Data labeling. Observations marked back_calculated were computed retrospectively under point-in-time inputs and are not a live track record. Observations marked preliminary have not completed the review window that would make them official. Observations marked delayed were calculated on inputs past the staleness threshold. Observations marked revised were restated through a correction batch, and not_calculated days are permanent gaps where a close could not be computed and the reason is recorded — never interpolated. The level series is a systems artefact, not a track record.
  3. Not investment advice. Nothing on these pages is an offer, recommendation, or investment advice. Index levels are informational calculations over on-chain data; a level is not a price at which any transaction could have been executed, and an index deletion is not a claim that a tracker could have exited at that level.
  4. No benchmark claim. PHIL10 is not administered under any benchmark regulation and no representation is made that it constitutes a benchmark within the meaning of any such regulation. No product may track it, and none does.

Philidor Labs · Index Services · Calculated from on-chain data · Append-only, versioned, content-hashed