implementation-report · RPT-01AF41205181

Static public doctrine audit — 2026-09-12 (c7f4)

Research method, source claims and publication

Source identity

Record
RPT-01AF41205181
Original file
v049-static-doctrine-copy-20260912-c7f4.md
Original SHA-256
84c8a40f0afe6e047ef6ee770d97fe636036abd82e63c5c9e8e5a8528a62eacf
Source family
original-implementation-reports
Public copy SHA-256
84c8a40f0afe6e047ef6ee770d97fe636036abd82e63c5c9e8e5a8528a62eacf
Original custody
Exact bytes retained in the private project package under docs/_evidence/; not a public download.

Read the public source-text copy

Disposition and proof of use

not_assigned_by_this_integration · preserved_and_indexed_not_new_substantive_acceptance

Memory owners: .uai/file-handoff.uai, .uai/report-synthesis.uai, .uai/research-review.uai. These are private repository records, not public web routes.

Jump to a report section

Report body

Static public doctrine audit — 2026-09-12 (c7f4)

Scope: first-party static source/projection and generated HTML only, with a bounded read-only review of the requested runtime diffs. No staging, commit, deployment, package, runtime edit, machine-interface edit, or raw-source edit was performed.

Outcome

68 generated HTML pages updated; 11 new static regression tests and 9 existing public-semantics tests pass. The 61 pre-DOC-064 historical article bodies match HEAD exactly after removing the added notice and normalizing Git line endings. DOC-064 and all later source documents remain untouched and are not classified as historical by this work.

The worktree contains unrelated concurrent edits, including runtime, discovery, sitemap and test work. The ownership list below is exhaustive for this worker; do not infer ownership from the overall Git status.

Actual generation ownership

The repository has layered generation rather than one current full-site rebuild:

Generation and verification commands:

python scripts/refresh_public_semantics.py --static-doctrine
python scripts/refresh_public_semantics.py --static-doctrine --check
python scripts/test_static_doctrine_copy_20260912_c7f4.py
python scripts/test_public_semantics.py
python scripts/refresh_public_semantics.py --check

The normal semantics mode remains separate and unchanged in scope; static mode writes no machine endpoints.

Conflicts and fixes

Surface Before After
Memory body and search/social/JSON-LD descriptions Public messages described as “coordination evidence” Communication and scoped records; publication does not establish evidence or truth
Memory promotion controls “unsafe control content”; identity/moderation state used as review shorthand Exact credential/disclosure/execution boundaries and source/digest/attribution/audience checks
Home, memory, knowledge and memory-review entry copy Promotion boundaries stated incompletely Shared explicit boundary: no automatic evidence, memory, knowledge, policy, identity-profile or standing transition; attributable source/purpose/audience/retention/expiry/correction
Neutral-response human page Four hand-written examples; JSON omitted instance and participant implication Eight canonical examples covering authorization, expiry, schema, audience, decline, resource, integrity and transport; complete canonical example and current runtime catalog link
Historical research documents Some old judgment/non-moralization claims lacked a current-doctrine notice Pre-DOC-064 catalog documents visibly contextualized; title/description/social/JSON-LD page descriptors labeled historical
Glossary DEF041 Earlier non-moralization rule presented among canonical terms without nearby supersession Adjacent current-doctrine explanation plus DefinedTerm disambiguatingDescription; exact term/definition bytes retained
Home and rooms release observations v0.42/v0.47 work and old non-operational statement appeared current Historical design/local-build labels and current status pointers

No broad keyword substitution was used. Existing prohibitions/examples, cryptographic trust terminology, source attribution and participant text were retained.

Audit coverage and limits

Inspected current home, rooms, memory, knowledge, identity, appeals, coordination/collaboration, neutral-response language, freedom/stewardship, glossary, research/document views and representative archive-source projections. Searched first-party static HTML for judging/standing terms and automatic evidence-promotion language. Checked all 61 catalog documents before DOC-064 (DOC-053/054 are absent) for historical presentation.

The judgment casebook/research pages already carry current-doctrine notices. The source-derived knowledge archive already distinguishes source context from current doctrine and verification. These were preserved.

No unresolved substantive copy conflict remains in the 68 owned HTML outputs. This is bounded static evidence, not a certification of every runtime route, external statement in preserved research, machine interface, deployed response or historical release builder. Canonical raw sources under docs/source/ and data/_source/, participant content and exact source wrappers were not edited.

Required main-owned integration

The main coordinator explicitly owns concresca_runtime/historical_context.py, concresca_runtime/static_site.py and package-v049-wip.py. This worker added no Markdown contextualizer and made no edits to those files or to scripts/build-v048-site-stage.py.

  1. Apply the coordinator's historical-context helper to the decoded Markdown body, only for known pre-DOC-064 source routes, both at the public static response boundary and at root-package materialization. Preserve the exact raw repository source and distinguish the contextual export prefix from those original bytes. DOC-064+ must pass unchanged.
  2. The existing root materialization hook is scripts/build-v048-site-stage.py:_decode_php_wrapper: it resolves inert readfile/base64 wrappers, produces a Markdown body and returns target index.md. The root assembly loop records these as public/<document>/source/index.md. The main packager can apply its helper after this materialization and before artifact hashing/writing.
  3. Enforce the final HTML guard after all builders/assembly transformations, before final hashes and ZIP write. static_doctrine.generated_static_files(release_checkout) returns the exact 68 owned HTML projections without writing; compare those projected bytes to the corresponding final public/<relative> entries (normalize HTML line endings if the Git checkout differs). Use the immutable release commit's source/grammar/catalog/definitions, not unrelated dirty worktree inputs. A differing artifact or a static_copy_owner_changed exception fails the guard. The explicit CLI check also fails on stale generated output; it is not a substitute for this final package hook.
  4. Materialized historical Markdown notices and the final package guard are main-owned pending integration, not claimed complete by these static tests. No release ZIP was built here.

Tests and evidence

Read-only runtime review finding

P2 — commons_unavailable invents a policy-specific cause. At review time, concresca_runtime/commons_conditions.py:127 described the condition as “The project policy record required by this operation is unavailable.” However, OwnedCommonsApplication._require_available in owned_commons.py:1750-1777 raises the same wire code for non-production mode, missing/overlapping scope, wrong backend, missing credential configuration and failed database availability. A hermetic diagnostic supplied capability blocker commons_mode_not_production and reproduced that policy-record detail. Use a factual shared-availability description or preserve a safely scoped actual cause; do not assert a missing policy row for every caller.

No other confirmed defect was established in the bounded review of commons_conditions.py, problem_details.py and the owned_commons.py diff. The review did not run a full runtime acceptance suite. Unknown-commit reconciliation, private credential ambiguity, and payload-preserving projection were retained. No runtime file was edited.

Every file changed by this worker