Reconcile the current-main CMP adoption ledger before upstream sync

[alexeygrigorev]

Parent/reference: #145 (closed controlled CMP synchronization)
Consumer/related: #290 (static Datamailer source inventory), #50 (Datamailer retirement and Relay purpose wiring)

Product outcome

Restore one truthful, deterministic adoption ledger for the adopted Course Management Platform before any future CMP source synchronization. The ledger must describe the exact reviewed target tree and its intentional target-owned overlays, while keeping the current CMP pin unchanged. A no-op sync against that pin must be an accepted, repeatable proof rather than a blocked or silently regenerated report.

This is a source/adoption-evidence repair. It does not adopt the current CMP main, copy its campaign changes, or authorize any runtime, provider, migration, or deployment behavior.

Evidence boundary

• Website authority for this audit: origin/main face8e4808d65afbf0374d1ced7a88079950d663, tree c8496fb2a37529eb7c63e1eeb9e35e29a2aadada.
• The active shared current-main-derived worktree is redesign-iterations ff30774e9cde310a24abf61229c22cbd5a811186, tree 62e1656881aa7f9de8ab703c14f5a29049c2097d. Its three additional local design edits are identified below; they must not be silently promoted to website main by this issue.
• Website source pin remains CMP 98a235283904b4ef9ad29e196298540756cf1bcc (source-pin.json SHA-256 71145a6b28ed0ea89e984b9c6bda5151549d3597d37c75d4c6b369ef4fbaf1b3). The copied-files ledger SHA-256 is 31a0ef0439b68477b905d9d084bfe9c0998dd8a7168ee3c99bc98d72542ea6e2; the integration-patched-files ledger SHA-256 is db285bab70aef0035235b68e17e63be55e4c133cfe514e0bd9026e2b76ac6c95.
• CMP main was independently audited at 6d3cc0e0e0b9a1d28be1a8d53209555269e1d3fe, tree 45f1f8e2548b25e3ef89d44b2370efbb647ff088, parent e49c575ce3f567cc14a8f66a89d3ccaca3bfe2f4. This issue does not move the website pin to it.
• The current sync dry-run fails closed on the first stale overlay row (accounts/templates/accounts/login.html); it must not be made green by ignoring the ledger or by running --apply.

Exact 15-row mismatch inventory

The current-main-derived audit compared each target overlay byte/size with the recorded integration-patched-files.tsv row. The observed set is exactly these 15 rows. Thirteen rows are already divergent on the authoritative face8e4 tree; the three rows marked redesign-only exist only in the unmerged local ff30774 worktree. The engineer must freeze the final accepted target identity before editing any ledger and preserve that distinction.

destination	recorded bytes / SHA-256	observed bytes / SHA-256	source
accounts/templates/accounts/login.html	2923 / 5adbd00244ed6229086681516829697165b1373679a6de95136561e956c81b04	2940 / 531514cb41b8323f743af88201d9ea6b67cbaceb180e2bdce730f9ba9ebe8e07	face8e4
accounts/views/login.py	6194 / 55dfc487be5e9fe00692b1c5b0c081b5d13e577f02b9f9dcce4c6f7ffd162c62	6251 / 23254ef15aefdc971e88379521bebecc932c10ead95080f45b5729ed43a08fba	face8e4
courses/registration.py	4988 / 4728a96c0106ef54b1644f16766ea8f4a3b4e9a52057a3ea3e3e000d2b005204	6221 / 60cd215bd6aa67e0cb441cf1a98e074c29ebc1779002d3857965e78bfe8c7517	redesign-only
courses/templates/courses/course_list.html	20395 / 5d08f07d4ab9064aeb906cb9435f0b287cae248b9f5affda7e9eae82fe126854	20428 / c406b4ca288a398f45030fb4032acaed43bbc3f46c03953f7b0673970b69dd72	face8e4
courses/templates/courses/dashboard.html	21030 / 86fbb37b894ed50043d1e57dafeea0ea1a0833c100ea1b323faf8ae31e300122	23080 / 2bf2e39f0387bbba530421174a77ab718bd93c188c52e0d600f58b677c4d3f70	redesign-only
courses/templates/courses/enrollment.html	12012 / d0c11d76e714d789b9aca8921bde550ba7ee3685eb1683090bee17db5d5aa97b	12046 / a63afcf108030c3e0018ffb8a93b824382c1c365991bebd14944352b03084669	face8e4
courses/templates/courses/leaderboard_score_breakdown.html	17594 / f95e13c740a224cb6d793eb2b97e0ab82bdd2cf1d5f03d33da3ad200f636cb6d	17679 / f1b4101fdd1a37a0bcce462453fd2dad220c63f30aec49416f78f98a0b20cd16	face8e4
courses/templates/courses/register.html	20288 / 11e7b6438e3ad00f3dd236ec77ba318869e6e31b1ea9473f078326b66cdf2d5c	20305 / 53f08c7fcb0885aea4bd1e7495ddc06873b6f45a41d6a962cc13c81bf8d7ddcf	face8e4
courses/templates/projects/eval.html	6335 / 82b2b541c17aea908d5c329d40de1d3b6be7acb554444f237e18d165cbeb1747	6369 / 26c799672e9b6fe9a81b204514cc051c3be2ce1adcb8697a5c12697278f6af96	face8e4
courses/templates/projects/list.html	12065 / 2ce06f1fcdfa4049d18edd2faf092948bb888c8dac8ec40ed99cc51bc6fbcc25	12184 / de31659cc7b34ffca680e9210d75b6fda3c6eca684fecbd5c4a9a3cfe270b59c	face8e4
courses/templates/projects/list_all.html	5004 / 77dc3a8c320a57da2e93a3f82bace8d9700ae031a06053621ae569ccde0d5ac2	5021 / f85cbb0c1afc0ab3c72edd1ecf8406e13d0d15018b5b6dc2054b7a1235222328	face8e4
courses/templates/projects/results.html	7549 / 9335a10a0988b7bd9b21d73439f412aea5afa43ba486d5cd0e8c059bf5b25cc6	7600 / b1203097d353e61f4c24ac28c94164b1623fd42b83e746308a3ebda2ca575f5b	face8e4
courses/templates/projects/submissions.html	6763 / ae283afbf471711dcec6332464d99b214f45fe97032ac8854682010b9a69dfb9	6780 / 02d1cc0d2553efdb64fb1fd737d0ca39500fd7ff6b52ea7e3d28a3ff521d762b	face8e4
courses/tests/test_dashboard.py	1379 / 563c875b43314b8e6862e7c0f578950e265c91ef014efa52c0135fe8148b91da	2185 / 2785f875ddae8f77172e2deaaf412381bfbdd32d4f7a96e150e96d8737dc6314	redesign-only
courses/tests/test_dashboard_integration.py	7214 / 8a1a98a29bc1192cab48a176363f0b3f80e9319d7b3b03271e989362e094d5e5	7283 / 4f8c907ecacf175a8fe33a5e06180ebb1930799f41443a0cef014bdff1b1c6a4	face8e4

Scope

• Freeze one exact website base/tree accepted by the orchestrator; do not infer that the local redesign worktree is main.
• Reconcile only the listed adoption/overlay rows, with a row-level reason and source/target ownership for each.
• Keep the CMP source pin at 98a235283904b4ef9ad29e196298540756cf1bcc and preserve the existing #145 mapping, deletion, migration, and overlay guards.
• Refresh only the necessary versioned adoption evidence and the deterministic report/manifest values required by the chosen target identity.
• Record the exact final target commit/tree, source pin, copied-files digest, patched-files digest, row inventory, and no-op report digest in the engineer/tester handoffs.

Explicit non-goals

• No source-pin update to CMP 6d3cc0e or any other moving ref.
• No sync_course_platform.py --apply, source copy, migration rewrite/squash, runtime behavior, template behavior, API/Studio change, Datamailer/Relay operation, provider/network call, protected-data access, CI rerun, commit, push, merge, or deployment in this issue.
• Do not absorb the three unmerged redesign-only rows into origin/main or overwrite user work.
• Do not use this issue to decide or implement EmailCampaign/campaign queue/from_email behavior; that is the separate decision issue linked from Wire approved messages through Relay and retire Datamailer #50.

Dependencies and ownership

• Add controlled upstream synchronization from CMP main #145 is the completed synchronization procedure and guard authority; do not reopen it.
• Pin the adopted Datamailer source surface in a static inventory #290 consumes a truthful adoption ledger and must not be accepted from the stale face8e4 evidence until this reconciliation is complete.
• Wire approved messages through Relay and retire Datamailer #50 consumes Pin the adopted Datamailer source surface in a static inventory #290's later source inventory but owns no ledger rewrite here.
• The future CMP campaign decision is intentionally separate from this evidence repair and from Decision: Choose CMP system-evaluation, peer-assignment, score, and notification behavior #149's four system-evaluation/peer-task/score/notification behavior packets.

Acceptance criteria

• PM records and freezes one exact target commit/tree; the 15-row inventory is reclassified against that identity, with every row accounted for exactly once and no local-only edit silently promoted.
• The adoption manifests record the actual target-owned overlay bytes for every intentional row, retain the pinned CMP source bytes/provenance, and contain no unreviewed path, duplicate, omission, or migration replacement.
• uv run --frozen python scripts/verify_course_platform_adoption.py passes against a clean checkout at CMP 98a235283904b4ef9ad29e196298540756cf1bcc.
• A fresh no-op uv run --frozen python scripts/sync_course_platform.py --source-ref 98a235283904b4ef9ad29e196298540756cf1bcc --source-checkout <clean-checkout> --report <redacted-report> --dry-run returns status=no_change, records the exact pinned/requested SHA, and reports zero changed, copy, conflict, and fatal paths without mutating tracked files.
• make migrations-check, adoption-focused tests, format/lint/type checks, and the verification plan selected for the changed evidence pass; the report classifies every component exactly once and documents screenshots as not applicable only if the graph proves no render inputs changed.
• Independent tester recomputes the plan from the frozen target identity, verifies the manifest/no-op report and every acceptance criterion, and posts a tester-final report before PM acceptance.
• PM posts ACCEPT only after the independent tester pass; the eventual focused commit uses Refs #297 and the orchestrator performs the normal local no-ff merge/push lifecycle.

Browser/test scenarios

This is non-rendering evidence work, so product screenshots are not applicable when the recomputed plan proves render inputs unchanged. The tester still runs the graph-selected repository-local tier and checks that the no-op sync/verifier cannot alter a rendered route. No provider, external CMP operation, or production browser session is permitted.

1. Run the no-op sync twice against a clean checkout of the pinned CMP commit; both reports and all tracked target bytes are identical.
2. Introduce a synthetic stale row, missing row, duplicate, wrong hash/size, source-pin drift, or dirty source checkout in an isolated fixture; the verifier fails closed with a bounded reason and no target mutation.
3. Confirm the final adopted course and account route manifests remain navigable/read-only under the existing target identity; no campaign or Relay behavior is exercised.

Required lifecycle

The candidate remains uncommitted after engineering. The independent tester must capture exact base/head/tree and report/artifact digests, then PM accepts or returns it. No source sync, deployment, or upstream CMP adoption is implied by a green no-op.

[alexeygrigorev]

Grooming Complete

Verdict: READY FOR ENGINEERING after the orchestrator freezes the accepted target identity; currently blocked on that identity and the release/base recovery.

The issue records the exact 15-row adoption/overlay mismatch observed in the current-main-derived audit, while distinguishing the 13 rows already divergent on authoritative face8e4 from the two rows present only in unmerged redesign-iterations ff30774. It does not authorize promoting the redesign worktree, changing the CMP pin, or applying a sync.

The narrow deliverable is a truthful row-level adoption ledger plus a fresh, immutable no-op report against the existing CMP pin 98a235283904b4ef9ad29e196298540756cf1bcc. The engineer must preserve #145's guards and prove adoption verification, no-op determinism, and the graph-selected verification envelope. Independent tester and PM gates remain mandatory under _docs/PROCESS.md; the final technical commit uses Refs #297 because this is evidence consumed by downstream adoption/retirement work.

Dependencies/consumers are explicit: #145 is the completed sync authority; #290 cannot consume stale adoption evidence; #50 consumes #290 later. The separate campaign product decision is #298, and #149 remains the independent four-family owner gate. No code, CMP/content repository, provider, CI, deployment, or worktree state was changed by grooming.

[alexeygrigorev]

Independent QA Review — #297 CMP adoption ledger reconciliation

Verdict: FAIL / return to engineering. I made no product-source edits,
commits, pushes, label changes, or PM acceptance. The candidate remains
uncommitted.

Frozen identity and plan

• Candidate worktree: /home/alexey/git/dtc-website/.tmp/issue-297-engineer
• Branch: issue-297-engineer
• Base/HEAD: face8e4808d65afbf0374d1ced7a88079950d663 /
  face8e4808d65afbf0374d1ced7a88079950d663
• Target tree: c8496fb2a37529eb7c63e1eeb9e35e29a2aadada
• Candidate status remained exactly the expected three paths:
  _docs/adoption/course-platform/integration-patched-files.tsv,
  scripts/sync_course_platform.py, and
  scripts/tests/test_sync_course_platform.py. git diff --check passed.
• Engineer plan: .tmp/verification-297/verification-plan.json, SHA-256
  b13da92df79c7500cb1dd9ff165e1ae13e37e17298330a87bc80649fb1165e5a.
• Independent tester plan:
  .tmp/verification-297-tester/verification-plan.json, SHA-256
  211c654cf392b41102da0acca990aaaac8b819ff04946f06e17c221e52da61cb.
• CI selection SHA-256 (identical engineer/tester selection):
  c884d9bd245c6997566dc4a1c76c45d1004b8a0d47d63d92a10d677328f9287a.
• Graph SHA-256: 05ffeee99244c9becd5e98e40be08188f1c5e2a61204daece984c3beb1d1bce6.
• Both plans select profile full, reason test_infrastructure, direct nodes
  surface.documentation/surface.scripts, and render.impact=false with
  screenshots not_applicable.

The plan handoff is stale. The engineer plan records source-tree manifest
digest 664b32e5bdc2078cf2173ec8a6bba5f7748790b38f00c6effcdeb4ca0e6cc0b9,
while the independent plan, recomputed from the frozen candidate, records
3c8eb6a80386ca8de7f2fd396b0839bf0d8583dfa03d1aeeeac671d4edf15f2f.
The retained status evidence shows .tmp/verification-297/status-before.txt
with only the ledger changed, followed by
.tmp/verification-297/status-before-noop.txt with all three candidate files;
the sync implementation and tests were added after the engineer plan was
created. Under _docs/PROCESS.md, this invalidates the engineer handoff and
requires plan regeneration before acceptance.

Fifteen-row reconciliation

Independent ledger/hash audit: .tmp/verification-297-tester/ledger-audit.log.
The candidate's patched ledger SHA-256 is
4264d37876e6cde9719d4a053164626efaba564bf1a6ba19219a8f6b364dc3e0.
Exactly 13 ledger rows changed relative to the target HEAD (rows 1, 2, 4–13,
and 15); each candidate size and SHA-256 matches the corresponding
face8e4 file. Rows 3 and 14 remain unchanged and also match face8e4.
No redesign bytes were copied into the candidate.

The issue text still has a material 2-versus-3 contradiction: its prose says
that two rows are redesign-only, but its table labels three rows as such:
courses/registration.py, courses/templates/courses/dashboard.html, and
courses/tests/test_dashboard.py. A direct read-only comparison with local
ff30774 confirms redesign-only bytes for all three rows. The candidate
correctly retains the face8e4 values for rows 3, 5, and 14 (row 5 is the
current target value despite the stale table label), but the issue's accepted
target wording has not been corrected. This criterion cannot be accepted until
PM rewords the inventory to account for every row exactly once.

CMP pin, adoption verifier, and no-op determinism

• Pinned CMP checkout .tmp/cmp-source-98a2352 was detached, clean, and at
  98a235283904b4ef9ad29e196298540756cf1bcc.
• uv run --frozen python scripts/verify_course_platform_adoption.py — pass:
  768 copied files and 4 target-owned compatibility shims.
• uv run --frozen pytest scripts/tests/test_sync_course_platform.py -q —
  10 passed.
• uv run --frozen pytest core/tests/test_course_platform_adoption.py -q —
  9 passed, 938 subtests passed.
• Two fresh runs of the exact pinned dry-run command both returned
  status=no_change, with the requested/current SHA exactly equal to
  98a235283904b4ef9ad29e196298540756cf1bcc and zero changed, copy, conflict,
  fatal, or follow-up paths. Reports were byte-identical, SHA-256
  ef60726afc2f217722f7bbcc3210b6cd9c3d7f5dabd7f9def6ef41f414d363c1; tracked
  target status was unchanged before and after both runs.
• Source pin SHA-256 remains 71145a6b28ed0ea89e984b9c6bda5151549d3597d37c75d4c6b369ef4fbaf1b3;
  copied-files ledger remains 31a0ef0439b68477b905d9d084bfe9c0998dd8a7168ee3c99bc98d72542ea6e2.
• The candidate diff is limited to the three expected paths and contains no
  EmailCampaign, campaign-queue, or from_email scope. No CMP/content source,
  apply sync, provider, or deployment operation was performed.

Independent gates

Gate	Result
uv run --frozen pytest ci/tests tests_ci -q	PASS — 607 passed
make lock-check	PASS
make database-portability-check	PASS
make security-check	PASS
make lint	PASS
make migrations-check	PASS — No changes detected
make django-check	PASS — OpenAPI, management parity, Django checks
make deployment-check	PASS
make verification-container VERIFY_CONTAINER_OUTPUT=.tmp/verification-297-tester/evidence/container-check.json	PASS — 13 assertions, artifact SHA-256 b3954adbc0731524e6d993d4c23d25d208c97a1d256118e79be796b7bb1f8647
uv run --frozen python -m ci.quality_contract --repository .	FAIL at terminology-check: known #253 frozen projection digest mismatches in podcasts.json, wiki.json, and wiki_search.json
make format-check	FAIL on six known baseline files; no candidate file is among them
make typecheck	FAIL on four known baseline errors in untouched content/event_speaker_bio_normalization.py (lines 505, 544, 550, 564)

The full quality contract stopped at the known #253 terminology mismatch. The
individual quality gates above were run independently to separate the known
baseline failures from this candidate; no #297-specific quality failure was
found.

Required browser policy

The recomputed plan marks Playwright as required=true with command
make test-playwright-smoke. A read-only census immediately before the browser
gate found another agent's full make test-playwright and Chromium process
group active in /data/agents/ai-shipping-labs/worktrees/worktree-agent-1551.
I did not start a competing browser run and did not kill or alter foreign
processes. Therefore the required independent Playwright component has no
acceptable tester evidence and the tester verdict cannot pass. Because the
plan proves no render inputs changed, product screenshots are correctly
not_applicable; this does not waive the required repository-local smoke run.

Disposition

Return to engineering. Regenerate the plan after all three candidate changes
are present, correct the issue's two-versus-three redesign-row wording through
PM, and run the required make test-playwright-smoke once the exclusive
browser slot is free. The candidate-specific adoption/no-op behavior is green,
but the stale plan, unresolved issue inventory wording, known #253/#261 quality
baseline, and absent required browser evidence preclude PM acceptance, commit,
merge, push, or deployment.

Evidence logs and artifacts are retained under
.tmp/verification-297-tester/.

[alexeygrigorev]

PM disposition — HOLD / return to engineering

The issue wording is corrected to consistently identify the three redesign-only rows shown by the table/evidence, while retaining the 13 current-main rows and the explicit no-EmailCampaign/campaign-queue/from_email scope.

Candidate-specific reconciliation is green: the 13 current-main ledger rows match the frozen face8e4 target bytes and hashes, no redesign bytes were promoted, the pinned CMP remains 98a235283904b4ef9ad29e196298540756cf1bcc, and the adoption verifier plus focused sync/adoption tests pass. Two fresh pinned dry-runs both returned status=no_change with exact SHA matching, zero changed/copy/conflict/fatal paths, and byte-identical reports (ef60726a…).

This remains on hold because:

• the engineer verification plan is stale: its source-tree manifest digest (664b32e5…) differs from the independently recomputed candidate digest (3c8eb6a8…);
• required make test-playwright-smoke evidence is missing because an exclusive foreign full-browser run occupied the slot; screenshots remain not_applicable only because the plan proves no render inputs changed;
• shared baseline gates remain red, chiefly the known Reconcile non-reproducible public projection after face8e4 #253 quality-contract projection digest mismatch (with the separately recorded baseline format/type failures).

Next gates are strict: engineer regenerates the plan from the frozen candidate; independent tester recomputes it, reruns the required smoke tier when the browser slot is free, and posts a final report; then PM re-evaluates and may accept only after those gates pass. No commit, merge, push, or deployment is authorized by this hold.

[alexeygrigorev]

Independent Tester Final — Issue #297

Disposition: FAIL / return to engineering. The candidate-specific adoption
reconciliation is green, but the required full-profile tester report is red on
shared baseline quality, Django, and browser gates. I made no product-source
edits, commits, pushes, labels, PM acceptance, or deployment actions.

Frozen candidate and refreshed plan

• Worktree: /home/alexey/git/dtc-website/.tmp/issue-297-engineer
• Branch: issue-297-engineer
• Base/HEAD: face8e4808d65afbf0374d1ced7a88079950d663 /
  face8e4808d65afbf0374d1ced7a88079950d663
• Target tree: c8496fb2a37529eb7c63e1eeb9e35e29a2aadada
• Worktree source-tree manifest: 3c8eb6a80386ca8de7f2fd396b0839bf0d8583dfa03d1aeeeac671d4edf15f2f
• Status remained exactly the three expected paths: the patched adoption
  ledger, scripts/sync_course_platform.py, and its focused test file.
  git diff --check passed.
• Refreshed independent plan:
  .tmp/verification-297-tester-refresh/verification-plan.json, raw
  SHA-256 a6c5f74442f0918c05de361b25c3915fbdcca7680eae12b138638a6afa50269b.
  Its canonical plan digest (the report plan_sha256) is
  3045d7c6ac3f49039131a1b13ee3dbbccbc05e2020dbaee463d083ecc22747de.
• Refreshed engineer plan raw SHA-256 is
  c9b001c3ebc2b86229086ddc5645869a39757befa47e3c9040cbc1b7c84a12ed; a
  semantic comparison of both plans (excluding creation time) matched,
  including the current source-tree digest. Selection SHA-256 is
  c884d9bd245c6997566dc4a1c76c45d1004b8a0d47d63d92a10d677328f9287a; graph
  SHA-256 is 05ffeee99244c9becd5e98e40be08188f1c5e2a61204daece984c3beb1d1bce6.
• The plan is full / test_infrastructure, with fresh container, Django,
  CI evidence, Playwright smoke, quality, and selector components. Compatibility
  and large-content invariants are outside the impact closure. Render impact is
  false, so product screenshots are not_applicable.

Ledger and scope checks

• The exact 15-row audit found 13 ledger replacements relative to face8e4
  (rows 1, 2, 4–13, and 15); every replacement size/hash matches the target.
  Rows 3 and 14 remain unchanged and match face8e4. Patched ledger SHA-256:
  4264d37876e6cde9719d4a053164626efaba564bf1a6ba19219a8f6b364dc3e0.
• Direct comparison with ff30774 identifies the three redesign-only rows as
  courses/registration.py, courses/templates/courses/dashboard.html, and
  courses/tests/test_dashboard.py; none of those redesign bytes entered the
  candidate. The issue body is now corrected to say three redesign-only rows
  and explicitly excludes EmailCampaign, campaign queue, and from_email.
  The engineer report still contains an old sentence saying “two” redesign-only
  rows; that handoff prose should be corrected before acceptance, although the
  current issue wording and candidate ledger account for all 15 rows.
• CMP checkout .tmp/cmp-source-98a2352 is detached, clean, and pinned at
  98a235283904b4ef9ad29e196298540756cf1bcc; tree
  22a4a9944f353bce1f7779c52cf3ec0fa56a7f32. Source-pin digest is
  71145a6b28ed0ea89e984b9c6bda5151549d3597d37c75d4c6b369ef4fbaf1b3 and the
  copied-files ledger digest is
  31a0ef0439b68477b905d9d084bfe9c0998dd8a7168ee3c99bc98d72542ea6e2.
• Adoption verifier: PASS — 768 copied files and 4 target-owned
  compatibility shims. Focused sync tests: 10 passed. Core adoption tests:
  9 passed, 938 subtests passed.
• Two exact pinned dry-run syncs both returned status=no_change, exact
  requested/current CMP SHA, zero changed/copy/conflict/fatal/follow-up paths,
  unchanged target status, and byte-identical report SHA-256
  ef60726afc2f217722f7bbcc3210b6cd9c3d7f5dabd7f9def6ef41f414d363c1.
• Candidate diff is limited to the three expected files and contains no
  EmailCampaign, campaign-queue, or from_email scope. No source copy,
  --apply, provider, network, migration, runtime, or deployment action was
  performed.

Independent gate results

Green candidate/control-plane checks:

• make test-ci — 607 passed (fresh tester run).
• make lock-check, make database-portability-check, make security-check,
  make lint, make migrations-check, make django-check, and
  make deployment-check — PASS.
• make verification-container — PASS, 13 assertions; artifact SHA-256
  b3954adbc0731524e6d993d4c23d25d208c97a1d256118e79be796b7bb1f8647.

Known base failures, with no changed candidate path involved:

• quality-contract-v1 — FAIL at terminology-check on the known Reconcile non-reproducible public projection after face8e4 #253
  frozen projection digest mismatches in content/public_projection/podcasts.json,
  wiki.json, and wiki_search.json.
• make format-check — FAIL on the six known untouched Restore face8e4 formatting, typing, and focused content gates #261 baseline
  files; make typecheck — FAIL on the four known untouched
  content/event_speaker_bio_normalization.py errors. No Reconcile the current-main CMP adoption ledger before upstream sync #297 file is in
  either failure set.
• make test-django-full — FAIL after finding 2,553 tests, at the known
  Restore face8e4 formatting, typing, and focused content gates #261 public-events fixture failure in
  content/tests/test_events_timeline.py::test_a_live_podcast_recording_keeps_the_mint_pill_the_index_gives_it,
  followed by the existing parallel-test traceback-pickling error. No Reconcile the current-main CMP adoption ledger before upstream sync #297
  path is implicated.

Required browser gate

A fresh process census immediately before the browser run found no active
Chromium/Playwright/pytest-browser process. I then ran the exact planned
command:

DTC_TEST_RUN_ID=tester-297-refresh make test-playwright-smoke

The run selected 30 tests and exited 2 after 205.11 seconds. Pytest reported
23 passed, 7 failed, 4 errors (teardown errors overlap failed test cases); the
flake-policy line reports attempted=30 passed=22 failed=8 skipped=0 rerun=0 quarantined=0 complete=0. Raw log SHA-256 is
6acea2e12d1e62dbe1c396fa23945505d6a05ffece7481b07ce3f5f894fa878a.

The failures are inherited public-surface/content behavior: missing
“Something to attend this week”, missing featured-event link
“AI Dev Tools Zoomcamp 2026 Course Launch”, and a 500 on /podcast?season=12
because a fixture resource is not HTTPS (ImproperlyConfigured), plus related
teardown timeouts. They do not touch any #297 changed path and are consistent
with the known #253/#261/base failure clusters. No browser process remained
after the run. Screenshots remain correctly not applicable because the plan
proves no render inputs changed.

Evidence envelope and disposition

All six required rerun components have tester-origin evidence envelopes under
.tmp/verification-297-tester-refresh/evidence/; the three not-applicable
components are classified by the plan. make verification-report-check
completed with exit 0 (envelope/report validation succeeded). The resulting
report is .tmp/verification-297-tester-refresh/verification-report.json, raw
SHA-256 04eaf489b1ee29fe93a5c0a6c29a8c9e46c70ee2d362e4434fd79e41b8e8e41a and
verdict failure; invalid_evidence=[].

Return to engineering. The CMP ledger reconciliation and deterministic no-op
are ready from a candidate-specific perspective, but PM must not accept,
commit, merge, push, or deploy this candidate while the required report is
red. Resolve or explicitly baseline the #253/#261 failures, correct the stale
“two” sentence in the engineer handoff, then regenerate the plan and repeat
the tester gates on the same frozen identity.

[alexeygrigorev]

PM disposition — HOLD

The independent tester recheck is recorded at issuecomment-5487419075; the corrected verification report has SHA-256 66addd9857e5b9719a1e684d89cdc1f427c62ebc6e026868576b17c4725324ea.

Candidate-specific reconciliation is green: exactly 13 current-main ledger rows changed and match frozen face8e4; the three redesign-only rows are courses/registration.py, courses/templates/courses/dashboard.html, and courses/tests/test_dashboard.py, and none was promoted. The CMP pin remains 98a235283904b4ef9ad29e196298540756cf1bcc. Two pinned dry-runs both returned status=no_change, exact requested/current SHA, zero changed/copy/conflict/fatal/follow-up paths, and byte-identical reports. make verification-report-check completed successfully with invalid_evidence=[].

This remains on hold: the full-profile report is red only on shared baseline gates—known #253 quality projection mismatches, #261 full-Django event-fixture/traceback failures, and the required smoke baseline (23 passed, 7 failed, 4 errors)—with no #297-specific defect identified. Acceptance, commit, merge, push, and deployment await #253/#261/#295 recovery and final green release gates. No lifecycle transition beyond this PM hold is authorized.

[alexeygrigorev]

Software Engineer Recovery Report — #297 on v7 rehearsal

Status: the #297 adoption-ledger lane is reconstructed and frozen as an
uncommitted rehearsal candidate. It is not tester-ready or PM-accepted, and no
commit, push, merge, deployment, CMP source change, or sync --apply was run.

Frozen identity and delta

• v7 staged input: face8e4808d65afbf0374d1ced7a88079950d663 with staged diff
  d51e35f9bcf6f2a820073e7ad938c5cb6095f959a96ce0082bc87574221d660 and staged
  tree ab68950b0655318f014e4e47db1bcf6a43247568.
• Candidate: .tmp/issue-297-reconstruct-v7, detached at face8e4, with all
  v7 staged bytes preserved and the Reconcile the current-main CMP adoption ledger before upstream sync #297 lane staged on top.
• Candidate staged diff: 5707277d4e8391871ecfc8998fdc250495ef441c6e5be1bd97f50e5ac4ce4223;
  staged tree d471a32d63eb4f5d6976baa1b00fdab1d8dea1d6.
• v7-to-Reconcile the current-main CMP adoption ledger before upstream sync #297 delta is exactly three paths, 73 insertions/deletions, with
  diff fingerprint 6aa946a29a5a868dc609034d60df3c5ba867c9b61c6138435626d9b026e1cca9:
  _docs/adoption/course-platform/integration-patched-files.tsv,
  scripts/sync_course_platform.py, and
  scripts/tests/test_sync_course_platform.py.
• git diff --cached --check and git diff --check pass. The full rehearsal
  index has 115 staged paths (the 112 v7 paths plus these three Reconcile the current-main CMP adoption ledger before upstream sync #297 paths).

Exact ledger scope

The patched ledger SHA-256 is
4264d37876e6cde9719d4a053164626efaba564bf1a6ba19219a8f6b364dc3e0 and has
exactly 13 row replacements against frozen face8e4:

accounts/templates/accounts/login.html, accounts/views/login.py,
courses/templates/courses/course_list.html,
courses/templates/courses/dashboard.html,
courses/templates/courses/enrollment.html,
courses/templates/courses/leaderboard_score_breakdown.html,
courses/templates/courses/register.html, courses/templates/projects/eval.html,
courses/templates/projects/list.html,
courses/templates/projects/list_all.html,
courses/templates/projects/results.html,
courses/templates/projects/submissions.html, and
courses/tests/test_dashboard_integration.py.

The three redesign-only rows remain excluded and retain the frozen face8e4
bytes: courses/registration.py,
courses/templates/courses/dashboard.html, and
courses/tests/test_dashboard.py. Their ff30774 redesign hashes are not in
the candidate. The CMP pin remains
98a235283904b4ef9ad29e196298540756cf1bcc; source-pin.json is unchanged
(SHA-256 71145a6b28ed0ea89e984b9c6bda5151549d3597d37c75d4c6b369ef4fbaf1b3),
and copied-files.tsv is unchanged (SHA-256
31a0ef0439b68477b905d9d084bfe9c0998dd8a7168ee3c99bc98d72542ea6e2). The
#297 delta contains no EmailCampaign, campaign-queue, or from_email scope.

Evidence

The clean-base #297 reconstruction remains the candidate-specific control at
.tmp/issue-297-engineer:

• uv run --frozen python scripts/verify_course_platform_adoption.py — pass,
  768 copied files and 4 target-owned compatibility shims.
• uv run --frozen pytest scripts/tests/test_sync_course_platform.py -q —
  10 passed.
• uv run --frozen pytest core/tests/test_course_platform_adoption.py -q —
  9 passed, 938 subtests passed.
• Two exact pinned dry-runs — both status=no_change, exact requested/current
  SHA, zero changed/copy/conflict/fatal/follow-up paths, unchanged tracked
  status, byte-identical report SHA-256
  ef60726afc2f217722f7bbcc3210b6cd9c3d7f5dabd7f9def6ef41f414d363c1.

The exact v7 reconstruction passes the #297 focused sync tests (10 passed),
but its full adoption verifier and adoption contract are blocked by other
unaccepted v7 lanes. The verifier reports the v7 #296 optional-signup edits
to accounts/templates/accounts/login.html and accounts/views/login.py,
plus #265 observability edits to
course_management/observability/{__init__.py,cloudwatch.py,cloudwatch_dashboard.py,events.py}
and data/tests/test_observability.py, which are not part of #297's 13-row
inventory. The v7 dry-run is correspondingly status=blocked (first fatal
path: accounts/templates/accounts/login.html) and did not mutate tracked
files. The adoption contract has 7 passing tests and 2 failures for those
cross-lane overlay/reference changes.

On the v7 candidate, the versioned plan is
.tmp/verification/issue-297-reconstruct-v7-engineer/verification-plan.json
(raw SHA-256
731f14176a3de6fdebdd443524a13b897386f2d1640fce91e30cae46f16a2827). It
records the exact worktree, 115 paths, full / cross_application, graph
05ffeee99244c9becd5e98e40be08188f1c5e2a61204daece984c3beb1d1bce6, source
tree manifest 4fa9e9a9e791bae619f29ff07717c65ce965409f9e1e973a23f602ec860fe341,
and render impact true. The independent tester must recompute this plan and
run the required full browser/screenshot envelope.

The v7 quality/control-plane checks are green: ci.quality_contract (all
selected targets, including test-ci 607 passed), lint, format (645 files),
typecheck (467 files), migrations, Django checks, deployment, security, and
lock checks. These checks do not waive the adoption verifier, independent
tester, or PM gates.

Remaining lifecycle gates

The v7 rehearsal cannot advance to tester/PM acceptance until the integrated
candidate's unaccepted #265/#296 overlay changes are separately reconciled or
removed by their owning lanes. #265 remains rehearsal-only pending the fresh
#32 accepted-slice and PM gate; #112 remains a human/nonmergeable gate. #297
itself still requires an independent tester report and PM ACCEPT. The
separate #262 Sponsors decision remains outstanding (REMOVE or RETAIN).