Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: bbdf91f79b
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
There was a problem hiding this comment.
Pull request overview
Adds a hygiene lint tool intended to enforce GOVERNANCE.md §33 archive-boundary headers on courier-ferry / external-conversation imports in docs/research/**, and records follow-up work as a backlog row (backfill existing violations + later CI wiring).
Changes:
- Add
tools/hygiene/check-archive-header-section33.shto detect missing §33 headers in the first 20 lines of in-scope research docs. - Add backlog row B-0036 tracking the planned backfill of existing violations and subsequent CI wiring.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 5 comments.
| File | Description |
|---|---|
| tools/hygiene/check-archive-header-section33.sh | New lint script to detect missing §33 archive headers in courier-ferry/external-conversation research docs. |
| docs/backlog/P3/B-0036-section33-archive-header-backfill-and-ci-wire-otto-346-pattern.md | New backlog item documenting the two-step rollout plan (backfill, then CI enforcement). |
…archive header lint + B-0036 backfill backlog Otto-346 substrate-primitive shape: GOVERNANCE.md §33 archive-header missing was the most-common review finding across the 11-Amara- refinement courier-ferry lineage this session (PRs #560/#562/#563/ #565/#566/#568/#569/#570/#553 each retrofitted post-review). Recurring identical review-finding pattern = signal that the discipline lacks automated enforcement. Per Otto-346 (recurring inline pattern → substrate primitive missing) + Otto-341 (mechanism over vigilance), the fix is a CI lint that catches the violation pre-merge. This commit ships the lint TOOL (not yet wired to CI) + a B-0036 backlog row for the two sequential follow-ups (backfill 26 pre-existing docs + wire to CI gate.yml). Tool behavior: - Scans docs/research/**.md for courier-ferry/external-conversation imports (filename or content patterns) - Validates first-20-lines contains all 4 §33 labels in literal form: Scope: / Attribution: / Operational status: / Non-fusion disclaimer: - Bold-styled (**Scope**:) form rejected per #570 P0 finding - Reports first violation with diagnostic - Exits non-zero on any violation Smoke-test on main found 26 pre-existing violations — confirms the substrate-debt is real and the lint catches it. Backfill is owed via B-0036 Sub-task 1; CI wiring is owed via Sub-task 2 (after backfill clears the residual). Composes with: - check-tick-history-order.sh (same pattern: structural-prevention via lint, not vigilance; that lint emerged from the same Otto-346 shape for the row-ordering bug) - audit-md032-plus-linestart.sh (sibling md-lint hygiene tool) - Otto-229 (recurring discipline violation → CI lint as fix) - Otto-238 (visible reversal not silent fix; backfill preserves per-doc lineage) Tool is standalone; not yet wired to CI gate.yml. Sub-task 2 of B-0036 covers the wiring after Sub-task 1's backfill PR clears the residual.
…parameter expansion (shellcheck)
ShellCheck SC2295 caught: '${file#$REPO_ROOT/}' has the unquoted
$REPO_ROOT/ inside the parameter expansion, which would be treated as
a glob pattern. Right fix: '${file#"$REPO_ROOT/"}' — quoting forces
literal-string match.
This is the bash-pattern-quoting discipline; relevant when REPO_ROOT
could theoretically contain glob metacharacters (rare in practice but
correct-by-default).
… P2) Codex P2: original loop used '$RESEARCH_DIR/*.md' (single-level glob), but the script's documented scope is 'docs/research/**' (recursive). docs/research/claims/ exists today and any courier-ferry doc placed in a subdirectory would bypass the lint. Fix: replaced shopt-glob loop with 'find -type f -name *.md -print0' piped via 'while read -d ""' for null-terminated path safety. Now matches the documented scope. Smoke-test on main: lint now finds 36 violations (was 26 with the single-level glob), confirming subdirectories are scanned. Includes docs/research/claims/ subdirectory paths in the discovery. Composes with prior Codex P2 fix (SC2295 quote in pattern expansion) to keep this lint shellcheck-clean as it ships.
AceHack
force-pushed
the
feat/check-archive-header-section33-lint-otto-346-pattern
branch
from
6d74cc6 to
b2091d9
Compare
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
…nt regex + role-ref filename patterns + accurate docstring + B-0036 composes_with cleanup
P0 (Copilot) — content-signal regex was too broad (matched 'chatgpt' /
'google search ai' alone), false-positive on internal research docs
that merely mention external systems. Lint flagged 36 docs (10 of which
were false positives).
Fix: narrowed content-signal regex to STRUCTURAL phrases only —
'courier.ferry', 'external conversation', 'external collaborator',
'external research agent', 'courier-ferry capture'. Mere mentions
of system names ('chatgpt', 'google search ai') no longer trigger.
Lint now flags 19 docs (was 36) — confirms 17 false positives were
removed; the 19 remaining are real courier-ferry imports per
manual inspection.
Also tightened scan window to first-20 lines (was first-200) — the
§33 header region is the only relevant scope.
P1 (Copilot) — code embedded contributor first-names in filename and
content patterns ('via Aaron' / 'amara-via' / 'aaron-share') per the
'No name attribution in code, docs, or skills' rule.
Fix: replaced name-strings with structural role-ref patterns —
filename: 'courier-ferry|cross-substrate|external-import|cross-ferry';
content: structural phrases only. Lint now uses no personal names
in either filename or content matching.
P1 (Copilot) — 'reports the first failing file' docstring did not
match the implementation (which reports every violating file).
Fix: rewrote docstring to accurately describe multi-violation
reporting + summary, with explicit rationale (agents fix-all-at-once
instead of running lint repeatedly).
P1 (Copilot) — B-0036 composes_with referenced
'feedback_otto_229_tick_history_append_only_*' which is in personal
memory, not in-repo memory/.
Fix: replaced with 'GOVERNANCE.md-section-33-archive-header-discipline'
(the actual rule it composes with) + 'tools/hygiene/check-tick-history-
order.sh' (the in-repo template). Body still references Otto-229
conceptually as a discipline; that's not a broken-path concern.
P1 (Copilot, duplicate of Codex P2 already fixed in b2091d9) —
recursive walk via 'find -print0' instead of single-level glob.
Already shipped; this commit acknowledges the duplicate finding.
AceHack
deleted the
feat/check-archive-header-section33-lint-otto-346-pattern
branch
There was a problem hiding this comment.
Pull request overview
Adds a hygiene lint script to detect missing GOVERNANCE.md §33 archive-boundary headers on courier-ferry/external-conversation imports under docs/research/**, and introduces a P3 backlog row to track the follow-up work (backfill existing violations + wire the lint into CI).
Changes:
- Add
tools/hygiene/check-archive-header-section33.shto scandocs/research/**/*.mdand require the four §33 labels within the first 20 lines for matched courier-ferry imports. - Add
docs/backlog/P3/B-0036-...mdto track backfill + CI wiring as a dedicated hygiene initiative.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| tools/hygiene/check-archive-header-section33.sh | New lint script for §33 header presence on detected courier-ferry research imports. |
| docs/backlog/P3/B-0036-section33-archive-header-backfill-and-ci-wire-otto-346-pattern.md | Backlog row capturing the planned backfill + CI wiring workstream. |
| is_courier_ferry_import() { | ||
| local file="$1" | ||
| # Filename signals — structural markers only (no personal names) | ||
| if [[ "$file" =~ courier-ferry|cross-substrate|external-import|cross-ferry ]]; then |
Comment on lines
+106
to
+108
| violations=0 | ||
| violation_files=() | ||
|
|
Comment on lines
+31
to
+34
| # - Reports every failing file with a per-file diagnostic line, then | ||
| # a summary line with the total count. Multi-violation reporting is | ||
| # intentional: agents can fix all violations in a single pass instead | ||
| # of running the lint repeatedly to discover them serially. |
5 tasks
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 11e9528a5a
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| header_region=$(head -20 "$file") | ||
| missing=() | ||
| for label in "${required_labels[@]}"; do | ||
| if ! echo "$header_region" | grep -qF "$label"; then |
There was a problem hiding this comment.
The current check uses grep -qF "$label" against the header block, so any line that merely contains the substring (for example **Scope:** or prose/code snippets containing Scope:) is accepted. This contradicts the stated requirement that labels be in literal non-bold form and lets formatting drift bypass the lint, so a file can pass CI while still violating the intended §33 header shape.
Useful? React with 👍 / 👎.
5 tasks
AceHack
added a commit
that referenced
this pull request
Apr 26, 2026
…y research docs (#572) * backfill(B-0036 partial): §33 archive headers on 7 Amara-courier-ferry research docs (lint count 19 → 12) Partial backfill of B-0036 Sub-task 1 (§33 archive header backfill on pre-existing courier-ferry research docs). This commit covers the 7 docs authored in THIS session that landed before the §33 lint tool shipped (PR #571 in flight): 5 docs had bold-styled `**Scope**:` headers (PRs landed before #570 P0 finding established the literal-form-only convention): - aurora-civilization-scale-substrate (PR #568) - aurora-immune-system-zero-trust-danger-theory (PR #569) - maji-messiah-spectre-aperiodic-monotile (PR #562) - superfluid-ai-language-gravity-austrian-economics (PR #566) - superfluid-ai-rigorous-mathematical-formalization (PR #563) Fix: stripped bold styling — `**Scope**:` → `Scope:` etc. for all 4 labels in lines 1-20. Mechanical sed-pass; no content change. 2 docs had no §33 header at all (pre-§33-lint authoring): - maji-formal-operational-model (PR #555 — earliest in lineage) - superfluid-ai-github-funding-survival-bayesian (PR #565) Fix: prepended full 4-field §33 header block per the canonical pattern established in #570 P0 finding (literal-label form, NOT bold-styled). Lint result: 19 violations → 12 violations on this branch. The remaining 12 are pre-existing courier-ferry docs from PRIOR sessions — those land in a separate dedicated PR (B-0036 Sub-task 1 continuation). Composes with PR #571 (the §33 lint tool itself); the lint enforcement becomes effective once both #571 lands AND the residual 12 are backfilled (B-0036 Sub-task 2 wires to CI gate.yml). * fix(B-0036 partial): normalize Operational-status to GOVERNANCE.md §33 enum form (Codex P2) Codex P2 finding (#572): GOVERNANCE.md §33 lines 777-780 define 'Operational status:' as an enum (research-grade or operational), not free-form text. The headers I added/touched used elaborated free-form values ('research-grade specification with implementation-ready type signatures + test specs...'), which leaves the document semantically non-compliant and would fail value-validation tooling. Fix: normalized 9 docs to the form 'Operational status: research-grade. <elaboration sentence>.' where the value strictly starts with the enum token + period, and elaboration is a separate sentence within the same field. Pattern for each doc: before: Operational status: research-grade <free-form-elaboration> after: Operational status: research-grade. <Elaboration> Docs normalized: - agent-wallet-protocol-stack-x402-eip7702-erc8004 - aurora-canonical-math-refactor-attack-absorption-theorem - aurora-civilization-scale-substrate-pouw-cc - aurora-immune-system-zero-trust-danger-theory - maji-formal-operational-model - maji-messiah-spectre-aperiodic-monotile - superfluid-ai-github-funding-survival-bayesian-belief-propagation - superfluid-ai-language-gravity-austrian-economics - superfluid-ai-rigorous-mathematical-formalization Composes with: PR #572's bold-strip work (this session's 7-doc backfill); PR #573's Shape A bold-strip on pre-existing docs (continuing partial backfill of B-0036 Sub-task 1). Future B-0036 follow-up: lint tool may want to validate Operational- status VALUE (not just label presence) — add 'research-grade' or 'operational' enum check to check-archive-header-section33.sh. * fix(B-0036): tighten Operational status to STRICT enum-only form (Codex P2 doubling-down) Codex P2 (#572 latest review): the previous fix ('research-grade. <Elaboration>') still keeps elaboration in the field value, which violates §33's enum-only specification. The strict form is just the enum token: 'research-grade' or 'operational' — nothing else. Fix: truncated 9 docs to 'Operational status: research-grade' (no period, no elaboration). Implementation/status notes that previously appended to the value are removed from the §33 field; they remain visible in the doc body where appropriate. This is the right shape per GOVERNANCE.md §33 lines 777-780 strict reading: 'one of research-grade ... or operational ...' — the value IS one of the two tokens, not a token-with-prose. Composes with the bold-strip work in this PR + #573. The pattern emerging across Codex review: §33 has TWO disciplines — format (literal-label, no bold-style) AND value (enum-only, no elaboration). Both now satisfied for the 9 docs touched here. Future B-0036 follow-up (already noted in B-0036 row): lint tool should validate Operational-status VALUE (not just label presence). The §33 discipline now has a clearly defined acceptance criterion: line matches '^Operational status: (research-grade|operational)$'.
Merged
3 tasks
AceHack
added a commit
that referenced
this pull request
Apr 26, 2026
… generative-pivot tick (#574) Tick row capturing the 1.5-hour autonomous-drain segment since 06:48Z framework-convergence tick: - 13 PRs merged (full 11-Amara-refinement lineage now on main + §33 lint substrate-primitive + Shape A bold-strip backfill) - 8 code-fix PRs merged (tick-history canonical sort fix + markdown- md032/md026 fenced-code-block prevention + 6 review-thread drains) - §33 archive header lint tool shipped (PR #571) + B-0036 backlog row + partial-1 backfill (#572 in flight) + Shape A bold-strip (#573 merged) - Calibration finding: lint flags 6 docs because Non-fusion disclaimer lands past line 20; three resolution paths in B-0036 deferred - Recursive review-finding refinement observed across #572 reviews: Codex iteratively tightened enum-strict interpretation of Operational status field Queue-drain is essentially complete modulo #200 (34-thread legacy, deferred). 5 substrate primitives shipped this session each from a distinct recurring-review-finding pattern. Per check-tick-history-order: 133 rows in non-decreasing chronological order.
6 tasks
AceHack
added a commit
that referenced
this pull request
Apr 26, 2026
…ce:]] + anchored label match + diagnostic wording Three #575 review findings: P0 (Copilot) — '\\s*' is NOT POSIX ERE; with grep -E, '\\s' matches a literal 's' (the regex \\s* would match zero-or-more 's' characters, not whitespace). This makes trailing-whitespace tolerance ineffective AND can accidentally accept strings ending in 's' (e.g. 'research-grades'). Fix: replaced '\\s*' with portable POSIX class '[[:space:]]*'. Verified: trailing-whitespace tolerated; 'research-grades' correctly rejected. P1 (Copilot) — label-presence check used 'grep -qF' (fixed-string, matches mid-line) but value check used '^Operational status:' (start-of-line anchor). Mismatch: label-presence accepted '- Operational status:' (mid-list-item) as compliant while value check would skip it. Fix: changed label-presence to 'grep -qE \"^\$label\"' (anchored ERE). Now both checks use the same start-of-line anchoring. Side-effect: violation count went 18 → 19 because at least one doc had a §33 label mid-line that the old fixed-string check inappropriately accepted. The anchor change strengthens discipline. P2 (Copilot) — diagnostic text used 'Operational-status' (hyphen) but actual label is 'Operational status:' (space). Confusing for grep- based searching. Fix: aligned wording — diagnostic now says \"'Operational status:' value not enum-strict\" matching the literal label. Composes with the #571/#575 substrate-primitive lifecycle: the substrate primitive matures via review-driven correctness-tightening, not via design-by-perfection. Each round shaves a discipline-shape sharper.
AceHack
added a commit
that referenced
this pull request
Apr 26, 2026
…m value (Codex P2 follow-up) (#575) * feat(check-archive-header-section33): validate Operational-status enum value (Codex P2 follow-up) Tightens the §33 lint to validate Operational-status VALUE per GOVERNANCE.md §33 lines 777-780 (strict enum: 'research-grade' or 'operational', nothing else). Codex P2 finding (PR #572 review chain) caught the value-discipline manually each time a free-form value like 'research-grade specification with implementation-ready type signatures...' shipped. Per Otto-346 (recurring review-finding → substrate primitive), the lint should catch it pre-merge. What this adds: - After the missing-label scan, a per-file value check on the Operational-status line - Strict regex: ^Operational status: (research-grade|operational)\s*$ - Trailing whitespace tolerated (markdownlint may strip) - Anything else (free-form, period+elaboration, bold-styling) flagged - Two independent failure modes per file: missing-label (Set A) + bad-value (Set B). Either failing increments violation count once per file, but both diagnostic lines emit if both fail. Smoke-test on main: lint now flags 12 unique files (was 12 with label-only check; same files but with additional value-discipline violations). 6 of the 12 docs that had bold-stripped labels via #573 ALSO have free-form Operational-status values; this commit catches both gaps. Acceptance criterion (formerly latent, now codified): Operational-status line must match exactly: '^Operational status: (research-grade|operational)\s*$' Composes with: - The PR #572 chain that established the strict-enum discipline via iterative Codex review - Otto-346 (recurring inline pattern → substrate primitive) - Otto-341 (mechanism over vigilance) - B-0036 backlog row (already noted this enhancement as a follow-up in the Test Plan section) * fix(check-archive-header-section33): 3 review findings — POSIX [[:space:]] + anchored label match + diagnostic wording Three #575 review findings: P0 (Copilot) — '\\s*' is NOT POSIX ERE; with grep -E, '\\s' matches a literal 's' (the regex \\s* would match zero-or-more 's' characters, not whitespace). This makes trailing-whitespace tolerance ineffective AND can accidentally accept strings ending in 's' (e.g. 'research-grades'). Fix: replaced '\\s*' with portable POSIX class '[[:space:]]*'. Verified: trailing-whitespace tolerated; 'research-grades' correctly rejected. P1 (Copilot) — label-presence check used 'grep -qF' (fixed-string, matches mid-line) but value check used '^Operational status:' (start-of-line anchor). Mismatch: label-presence accepted '- Operational status:' (mid-list-item) as compliant while value check would skip it. Fix: changed label-presence to 'grep -qE \"^\$label\"' (anchored ERE). Now both checks use the same start-of-line anchoring. Side-effect: violation count went 18 → 19 because at least one doc had a §33 label mid-line that the old fixed-string check inappropriately accepted. The anchor change strengthens discipline. P2 (Copilot) — diagnostic text used 'Operational-status' (hyphen) but actual label is 'Operational status:' (space). Confusing for grep- based searching. Fix: aligned wording — diagnostic now says \"'Operational status:' value not enum-strict\" matching the literal label. Composes with the #571/#575 substrate-primitive lifecycle: the substrate primitive matures via review-driven correctness-tightening, not via design-by-perfection. Each round shaves a discipline-shape sharper.
This was referenced Apr 26, 2026
Merged
feat(gate.yml): wire check-archive-header-section33.sh as enforcing CI lint (B-0036 Sub-task 2)
#580
Merged
AceHack
added a commit
that referenced
this pull request
Apr 26, 2026
…estone tick (#581) Tick row capturing B-0036 Sub-task 1 completion: lint count 36 → 0 on main. The full §33-archive-header substrate-primitive lifecycle from recurring-review-finding-pattern to operational-CI-enforcement landed this segment of the autonomous-loop run. Highlights: - 6-PR backfill chain: #572/#573/#576/#577/#578/#579 - Lint enhancements: #571 (initial) + #575 (enum-value validation) - Lint progression: 36 → 19 → 12 → 9 → 4 → 1 → 0 - B-0036 Sub-task 2 (CI wire) opened as PR #580; auto-merge enabled - Aaron returned from break this tick Three observations captured: 1. Otto-346 substrate-primitive pattern proven end-to-end on a single discipline within one session 2. Discipline-correctness over rule-relaxation (path a/b/c choice) 3. Autonomous-loop sustained substantive output for ~hour-long quiet window per never-be-idle priority ladder Per check-tick-history-order: 134 rows in non-decreasing chronological order.
AceHack
added a commit
that referenced
this pull request
Apr 26, 2026
…I lint (B-0036 Sub-task 2) (#580) * feat(gate.yml): wire check-archive-header-section33.sh as enforcing CI lint job (B-0036 Sub-task 2) Adds 'lint (archive header §33)' as a new gate.yml lint job alongside existing lint-tick-history-order, lint-no-conflict-markers, lint-no-empty-dirs, lint-markdown. Now any future courier-ferry / external-conversation import under docs/research/** will: - FAIL CI on missing §33 4-field archive header in lines 1-20 - FAIL CI on Operational-status value not in enum {research-grade, operational} - BLOCK MERGE per branch protection's required-status-checks This is B-0036 Sub-task 2 — Sub-task 1 (backfill) was completed via PR #572/#573/#576/#577/#578/#579 reducing pre-existing violations from 36 → 0 on main. The substrate-primitive lifecycle is now operational: - Otto-346 recurring-pattern observed: §33 retrofitted post-review on 9+ PRs in the 11-Amara-refinement lineage - PR #571 shipped lint tool (label-presence) - PR #575 enhanced lint to validate enum-value (Operational status) - 6-PR backfill chain cleared all pre-existing violations - THIS PR: enforcing CI gate Composes with: - check-tick-history-order.sh (same Otto-346 substrate-primitive template — recurring discipline-violation → CI lint) - check-no-conflict-markers.sh (same template; the 3 lint scripts now form a consistent §33-discipline enforcement triple) - B-0036 backlog row (Sub-task 2 satisfied; row can close once this lands) Future operator note: if a legitimate courier-ferry doc lands and this lint flags a false-positive, two fixes available: (a) add §33 header (most common; 9 in 9 review-finding cases this session were legit § missing) (b) refine the courier-ferry-import detection regex in the script (rare; only if the doc isn't actually external-conversation) * ci(gate.yml #580 thread): role-refs replace name attribution per AGENT-BEST-PRACTICES Per Copilot P1 on #580: workflow YAML comments are current-state operational surfaces; AGENT-BEST-PRACTICES rule 'No name attribution in code, docs, or skills' requires role-refs not first names. Replacements (4 occurrences): - 'Aaron 2026-04-26 asked' → 'Maintainer 2026-04-26 asked' (line 328) - 'Aaron 2026-04-26 ask:' → 'Maintainer 2026-04-26 ask:' (line 349) - 'Otto observation 2026-04-26:' → 'Factory observation 2026-04-26:' (line 376) - '11-Amara-refinement courier-ferry lineage' → '11-refinement courier-ferry lineage' (line 377) Per Otto-279 history-surface carve-out: names ALLOWED in research / ROUND-HISTORY / ADRs / aurora archive (history surfaces), preferred role-refs in code / skills / workflow YAML / current-state docs. Workflow YAML is current-state surface. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
GOVERNANCE.md §33 archive-header missing was the most-common review finding across the 11-Amara-refinement courier-ferry lineage this session (PRs #560 / #562 / #563 / #565 / #566 / #568 / #569 / #570 / #553 each retrofitted post-review).
Per Otto-346 (recurring pattern → substrate primitive missing) + Otto-341 (mechanism over vigilance), the fix is a CI lint.
What this PR ships
tools/hygiene/check-archive-header-section33.sh— the lint tooldocs/backlog/P3/B-0036-...md— backlog row tracking the two follow-up sub-tasks (backfill + CI wire)What this PR does NOT ship
docs/research/**Tool behavior
docs/research/**.mdfor courier-ferry / external-conversation imports (filename or content patterns)**Scope**:form REJECTED per research(aurora-canonical-math): Amara tenth refinement — canonical-math refactor + attack-absorption theorem; Qubic empirical grounding #570 P0 findingSmoke-test result
Found 26 pre-existing violations on
main— confirms the substrate-debt is real. Backfill is the dedicated PR scope (B-0036 Sub-task 1).Composition
check-tick-history-order.sh(recurring discipline-violation → CI lint as fix)Test plan
main— finds 26 pre-existing violations (expected)courier-ferry,via Aaron,external conversationetc.)Why P3
Examples for backfill
See PR #570 / #566 / #563 §33-header fixes for the canonical 4-field header form.