Skip to content

integration 2026-10-14: nineteen independently reviewed PRs combined for one merge - #329

Merged
MendixMau merged 100 commits into
masterfrom
integration/2026-10-14
Oct 14, 2026
Merged

integration 2026-10-14: nineteen independently reviewed PRs combined for one merge#329
MendixMau merged 100 commits into
masterfrom
integration/2026-10-14

Conversation

@MendixMau

@MendixMau MendixMau commented Oct 13, 2026

Copy link
Copy Markdown
Owner

Direct lane

What changed and why (one paragraph):
This branch combines 19 PRs. Each one was reviewed by a fresh agent that didn't write it, per the new merge rule in #316, and only MERGE verdicts went in. Each --no-ff merge commit carries its Review: line and the conflict resolution it needed. The branch also carries one follow-up commit for #328, which fixes run-all.sh resolving fixture subjects under bin/lib/. That fixture was red on #328's own CI when #328 was merged into integration; the fix is green on CI and was reviewed separately. The same pattern was used before in #258. Merge order (first parent):

# What
#315 Trim: frees baseline word budget by moving text, dropping no rules
#316 Process: independent review before every merge; word budget is made by moving, not deleting
#318 Page-fidelity scorer matches displayed text
#317 Interaction-model skill for Stage 3 (part 1 of 3 of #291)
#321 PROOF-OF-LOOK per section; final-look not due before derived Stage 6
#307 graph-sweep/app-facts edge guard; per-BRD conformance ledgers
#325 Extraction-quality section; token-burn rates/windows/effort
#310 OQL rows by alias; negative-only span liveness; APP-DOWN; #302 credential fix
#323 Inbox promotions
#308 Lint gate reads the install receipt; reverted stock rules reported BLIND
#328 pipefail + grep -q false negatives (CI flake), plus the run-all follow-up
#319 test-stack-up verifies run --local; shared port/process helpers
#327 change-intake core cut to baseline + on-demand detail; CRLF-safe change log
#320 Follow-ups to #290/#292; Dispatching split into its own skill
#326 status: delivery rounds in existing-app mode
#312 Lint batch 4: SEC014, toolkit CONV009
#324 exec.sh change classes; shared stale-lock takeover; SEC014 note echo
#322 app-restart / test-scratch refuse to restart over a live MxTest after-startup
#314 doctor/sync: property-set check errors get their own exit code

Left out: #311 (on HOLD until a field run) and #309.

Field evidence: most instruments here have not had a real-project field run yet. Each PR body says so where that's the case, so treat field runs as owed after the merge.

  • No client data anywhere in the diff (leak guard clean, no denylist)
  • Size cap: exceeded by design. This is an integration bundle; each constituent PR was reviewed on its own.
  • Test tier: CI was green on every constituent PR head. On this branch, in a shallow clone: tests/run-tests.sh 33/0; tests/wave2/run-all.sh 102 fixtures, 2 failing before the fix(routing): pipefail + grep -q false negatives (CI flake) #328 follow-up. One was the bin/lib subject bug, now fixed. The other was test-harvest-learnings.sh, which needs full git history; CI has it. render-routing.sh --check passes, with baseline at 74,959 of 80,000 words.
  • CHANGELOG.md lines come from each constituent PR

Draft until the maintainer merges it.

🤖 Generated with Claude Code

https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw

Generated by Claude Code

claude added 30 commits October 13, 2026 10:06
@claude
Fix graph-sweep/app-facts catalog guard and per-BRD conformance ledge…
5412e82 
…rs (#259, #260)

graph-sweep.sh and app-facts.sh refused any catalog whose build_mode label
was not 'full'; mxcli v0.24+ writes 'source' for a complete catalog. Both now
check what they read: a non-empty refs table with at least one RefKind='call'
edge, and FAULT naming the mode and per-kind edge counts otherwise.

conformance-check.sh read only <Module>/coverage-ledger.md, which is an index
in the per-BRD layout. It now measures every <Module>/coverage-ledger/*.md,
writes one report per BRD, keys baseline pointers by BRD, documents
LEDGER_FILE (file or directory), FAULTs on a missing LEDGER_FILE, and names
an index as an index with the files its rows are in.

Fixtures (both layouts, fail on master's scripts):
tests/wave2/test-catalog-call-edges.sh (12 checks, 10 fail on master),
tests/wave2/test-conformance-per-brd-ledger.sh (17 checks, 17 fail on master).
Synthetic catalogs/ledgers; field runs on a v0.24 / per-BRD project are owed.

Reported by AlvinnYeo.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
fix(lint-gate): reverted-to-stock rules stop the run as BLIND, not a …
4a60e14 
…rise (#264); BUG-156 + preflight row 28 (#263)

#264: `mxcli init` re-seeds .claude/lint-rules/ and the toolkit rules lose their header.
lint-gate.sh never read .claude/.mxtk-lint-receipt, so the re-seed showed up as a rise on the
next exec (CONV010 26 -> 169 on a pages-only script), or sat for days while the baseline
absorbed stock counts.

- project-bin/lint-gate.sh: before linting, each rule in the receipt is md5'd; a mismatch with
  the header gone (or a missing file) is named, the remedy `<toolkit>/bin/sync-project.sh
  <project>` is printed, and the run exits 1 with `BLIND — lint rules reverted to stock`.
  lint-last.json gets verdict BLIND-RULES-REVERTED and no total (no stale total for
  metrics-ledger). --update-baseline is refused over reverted rules. A headered edit is left to
  the installer's LOCALLY MODIFIED report; no receipt means no check;
  LINT_RULES_REVERTED_OK=1 lints anyway, with the warning.
- project-bin/exec.sh: run_lint_gate maps that output to LINT_RC 3 / "LINT BLIND: rules
  reverted to stock (<files>) — run sync-project.sh; delta not attributable".
- Guard rules: the receipt is the installer's evidence, not lint-gate's own stamp (rule 6);
  the remedy (sync-project.sh / the installer) is never blocked and clears it — fixture T6
  (rule 7).
- tests/wave2/test-lint-receipt-revert.sh: golden receipt from the real installer run into a
  scratch project; 23/23 here, 11 fail with master's lint-gate.sh.

#263 (learned-rule half): BUG-156 in bug-logs/mxcli-bugs.md (bin/bug-lookup.sh BUG-156 and
"free room" find it) and STOP row 28 in skills/learned-mdl-preflight.md, quoting the issue's
exec error verbatim. Upstream draft prepared, not filed.

Field run: the receipt/installer half was field-checked by the maintainer on a copy of a
v0.25.0 project (#264 comment: installer restores conv010/data_change, md5s then match the
toolkit). lint-gate's new check has not yet run on a live project — fixture only.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
Merge origin/master into fix/lint-receipt-check
39e4466 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
e2e: read OQL rows by alias; liveness for negative-only span claims; …
faf196e 
…APP-DOWN between journeys

#154: helpers.js built OQL rows with Object.values(r) and oqlScalar read
rows[0][0]. mxcli v0.26.0 prints keys alphabetically, omits an all-null
column and, over a view entity, adds ID beside the alias, so columns
shifted and a view's count read as its ID. The brace counter ignored
strings, so a value holding ] or } returned null. Rows are now keyed by
the parsed SELECT aliases (SELECT order, explicit nulls), oqlScalar reads
the first alias, new oqlRows, string-aware slicing. Golden: the released
v0.26.0 binary's verbatim output against a stub admin API
(tests/wave2/fixtures/oql-json/); test-oql-rows.sh 15/15, 0/15 on master.

#150 remainder: a mustNotFire-only step with zero spans was INVALID on
every run. Rung 2 (now traceRung) waits the full window and requires a
declared spans.liveness span or any service span since the journey began
(otel.js liveness) before judging; the trace-negative mutant injects the
forbidden name after that check.

#285 harness half: probe baseUrl before each journey (APP-DOWN, not
walked) and after each (flag its FAIL rows). APP_PROBE=0 disables.
test-journey-liveness.sh 27/27 over the captured Jaeger trace.

No live-app field run yet.

Closes #154
Refs #150 #285

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
Lint Batch 4: SEC014 external-role writes, CONV009 reads maxObjects (#…
8885d01 
…268, #267)

SEC014 (warning, security tier): an external user role, named in
lint-config.yaml as `external_roles: "A, B"` (or "none"), that can write
all members, a state enumeration, an audit ...At/...By attribute, or the
association its own XPath scopes ownership on. A `_rule` finding when the
option is unset, given as a YAML list (mxcli passes it as None), names a
role that does not exist, or the catalog is not FULL.

CONV009: toolkit-owned copy of the stock rule, which hard-codes 15 and
ignores the maxObjects option its own generated config advertises. Reads
maxObjects (default kept at 15) and countLoopBodies; replaces the stock
file through STOCK-HASHES.txt (conv009 at v0.17.0/0.24.0/0.25.0 plus a
full v0.26.0 set, captured with `mxcli init` from the official release
assets; test-stock-hashes.sh T4 passes live on all four).

Harness: get_option, permissions_for, attributes_for, user_roles,
role_mappings mocks. 26 cases in tests/lint-rules/test-batch4-rules.sh.

Field run (official mxcli 0.26.0, scratch copies of two models): stock
CONV009 5 -> 5 with maxObjects: 20; toolkit CONV009 5 / 10 / 6 at
15 / 10 / 20+loops. SEC014 1 finding with the external role configured,
7 with every role configured, matched by a SQL replay of the catalog.
Both rules also load on mxcli 0.17.0 and 0.24.0.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
e2e: validator accepts negative-only span claims; empty spans block i…
b5731a4 
…s a loud INVALID

#150 item 3: a spans block declaring neither ordered nor mustNotFire
(e.g. the array shape [{mf,status}]) is now recorded INVALID by
traceRung instead of being skipped silently.

examples/validate-journeys.js rejected a mustNotFire-only claim as
"NEVER checked", which contradicted the runner since the liveness change.
It now accepts it, validates the new spans.liveness field (read only for
a negative-only claim) and rejects a mustNotFire that is not a regex;
selftest 19 rules red, 3 clean cases green. skills/journey-examples.md
documents the liveness rule and the INVALID for an empty spans block.

Fixtures: test-validate-journeys-spans.sh 12/12 (8 fail on master's
validator); test-journey-liveness.sh 30/30.

Closes #150

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
feat(doctor, sync-project): wire mxcli fix property-sets into the upg…
cd187a4 
…rade path, report first; name a pre-fix model-stamp.sh

#295. mxcli up to v0.25 wrote units without Studio Pro's full property
set. mx check, the build and the runtime stay green, and Studio Pro's
merge/diff throws once it has saved such a document. No gate saw it.

- bin/doctor.sh --property-sets <dir>: read-only `mxcli fix property-sets
  --dry-run`, result plus mxcli version in .claude/.property-sets-receipt
  (gitignored). Exit 0 clean / 1 needs or unclassified / 2 not run /
  3 mxcli < v0.26.0. --if-stale skips when already run for this mxcli.
- bin/doctor.sh --fix-property-sets <dir>: the repair, only through
  bin/exec.sh --patch (snapshot, mxbuild gate, restore); refused on an
  exec.sh without --patch. doctor.sh is its own patch script.
- doctor --quick (run by exec.sh when the mxcli version changes) warns
  while the check is owed; a full doctor runs it.
- sync-project.sh runs the check with --if-stale (never the repair).
- Support is read from `mxcli fix --help`: `fix property-sets --help`
  exits 0 on v0.24/v0.25 too (cobra prints the parent's help).
- learned-detection-gaps.md row; platform-link.md "Before a Team Server
  merge" (fix property-sets, fix hashes, git note).

#236 remainder: known_fix_note gains the model-stamp.sh arm keyed on
GIT_WORK_TREE (patch 0004); test-model-stamp.sh T9 adds the unstaged
worktree `commit -a` and `commit -- app` cases.

Golden input: `--version`, `fix --help`, `fix property-sets --help` and
a not-found dry-run, captured from the official v0.24.0/v0.25.0/v0.26.0
linux-amd64 release binaries (never run against a model). The clean /
NEEDS / FAILED reports follow mxcli v0.26.0's source format strings.
tests/wave2/test-property-sets.sh: 43 passed here, 26 fail on master's
doctor.sh/sync-project.sh. Field run on a real model still owed.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
e2e: rewrite example.journey.json into the shape the runner reads
fc966d7 
The shipped example had drifted from journey-runner.js: name/role instead
of id/persona, a `value` seed instead of `sql`, page/expectPage/
expectText/click instead of actions + ui.ready/textPresent, spans as an
array of {mf,status} (now a loud INVALID), and SQL-style
<module>$<entity> queries where the runner sends OQL. validate-journeys.js
reported 3 problems on it; now 0. All 7 positive-control mutants are
supported on it; journey-runner.selftest.js (which falls back to it) and
test-journey-preflight.sh (back-compat cases) pass.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
chore(context): free baseline word budget without dropping rules
dc95126 
The always-on baseline tier counted by bin/render-routing.sh --check was
at 79,735 of its 80,000-word cap. This frees 3,772 words (now 75,964)
by moving narrative and setup-only text out of baseline files, keeping
every rule, STOP, denominator and gate cell in place, each rule still
paired with a one-line failure example plus a pointer to the full text.

- conversion-runbook.md, source-triage.md, interview-protocol.md: long
  incident narratives -> one line + pointer; originals verbatim in new
  on-demand process/baseline-incident-history.md (R1-R20, AR1, ST1-ST4,
  IP1-IP3). Runbook's duplicate open-questions status table now defers
  to interview-protocol.md section 4.
- agent-roles.md: the six template shapes (setup-only) moved to
  agents/README.md, which installers never copy (explicit MXTK_AGENTS).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
Merge remote-tracking branch 'origin/master' into chore/baseline-word…
f5ba690 
…-trim
@claude
process(merging): independent review before every merge
982e4e3 
A merge go is not a review. Before any PR merges, an agent that did not
write it reviews the diff and returns MERGE / MERGE-AFTER-FIX / HOLD on
value, regression risk after sync, correctness and context cost.
Builder self-reports and green CI no longer count.

Why: nine PRs merged on 2026-10-13 on builder reports and CI alone.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
process(context): word budget is made room for by moving, never deleting
8800dcb 
Rules, STOP rows, denominators and failure examples are protected.
Stories and examples move behind pointers in a separate, independently
reviewed trim PR; trim starts at 95% of the cap.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
Merge remote-tracking branch 'origin/master' into fix/lint-receipt-check
1729259
@claude
Merge origin/master into fix/graph-sweep-conformance
a8df6ad 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
e2e: review fixes — liveness window, stale otel.js guard, old spans s…
4110b92 
…hape named

- journey-runner.js: the negative-only liveness window starts at this
  journey's own start (LIVENESS_SKEW_MS = 2 s clock-skew margin), not 30 s
  earlier where the previous journey's spans counted.
- A project otel.js older than the runner (hand-edited, kept by sync)
  has no liveness(): the step reads INVALID naming the stale otel.js
  instead of a TypeError.
- spans as an array (old example shape) or {} gets an INVALID that names
  the shape; validate-journeys.js names the array shape too.
- Comment cited a non-existent fixture; "first cell" wording -> "first
  SELECT alias" (journey-examples.md, validate-journeys.js,
  full.journey.json); "16 rules" -> 19.
- CHANGELOG: callout that oqlScalar now returns the first SELECT alias.

Fixtures: test-journey-liveness.sh 36/36 (cases for the skew window, the
stale otel.js and the named shapes), test-validate-journeys-spans.sh 12/12.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
Merge origin/master into fix/e2e-oql-rows
e02d5f9 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
docs(journey-examples): rule count 19, without rewriting the example-…
7f73f8d 
…domain line

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
fix(trim): review follow-ups — dispatch obligation pointer, stale "be…
76075d8 
…low", ST1 id

Independent review of #315: obligations.tsv still pointed at agent-roles.md for
step 8a, which moved to agents/README.md; agent-roles.md said the example bodies
are "below"; the history header cited a non-existent section id.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
fix(lint-rules): the remedy lint-gate prints always clears it; delibe…
18c9a61 
…rate opt-out; prune stale receipt rows (#264 review)

Review of #308 (MERGE-AFTER-SMALL-FIX):

1. bin/lib/install-lint-rules.sh: a rule whose receipt row exists and whose `# mxtk-lint-rule:`
   header is gone is restored even when its hash is not in STOCK-HASHES.txt (an mxcli newer
   than the list, then init). It used to fall into the foreign branch -> LOCALLY MODIFIED, never
   written, so lint-gate stayed BLIND forever. Same test lint-gate uses; the found file is kept
   as <rule>.reseeded-<ts>. lint-gate's message adds the fallback
   `sync-project.sh <root> --upgrade-lint-rules all`.
2. Deliberate opt-out: .claude/lint-rules-disabled.txt. The installer never installs/restores a
   listed rule and prunes its receipt row; lint-gate skips it. lint-rules/README.md (txt002 row,
   CONV010 prefix note) now says list-then-delete, with an "Opting out of a rule" section.
3. The installer prunes receipt rows for rules no longer in MXTK_LINT_RULES /
   MXTK_LINT_RULES_CONFIGURABLE, so a retired rule is not "gone" forever.
4. skills/learned-mdl-preflight.md row 28 cut to trigger + rule + one-line failure (60 words,
   was ~230); the narrative stays in BUG-156.

Fixture tests/wave2/test-lint-receipt-revert.sh T9-T11 (INSTALLER= swaps the installer):
T9 uses a genuine stock body (toolkit entity_business_key minus header, md5 d66528c9 = the
v0.24.0/v0.25.0 stock hash in STOCK-HASHES.txt) under a scratch toolkit that does not list the
hash and whose rule has moved off stock. 38/38 here; 9 fail with master's installer, 21 with
master's installer + lint-gate.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
Add interaction-model skill: role homes, inboxes, page-parameter cont…
d55a4d4 
…ract before the first wireframe

Part 1 of 3 for #291. Stage 3 now writes design/interaction-model.md
after the role model is CONFIRMED and before the first wireframe: a
home per role, an inbox per BRD user task, and per screen how it is
reached, its page parameter / context object, its data view source,
nav entry and roles. Every wireframe opens with a wf-contract block
(Page parameter:, Reached from:, Drawn as role:), which page-fidelity
treats as annotation and the prototype hides behind Show bindings.

Wired: design-artifacts.md Step 2b + Step 3 contract + output tree,
conversion-runbook.md Stage 3 "Agent produces" (3 words; baseline is
now exactly at its 80,000-word budget), routing row (ondemand, Stage 3,
lead + architect) re-rendered.

Refs #291

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
Judge each PROOF-OF-LOOK by its own section's VALID AT; newest stamp …
50e69d0 
…by commit time (#155); no LOOK row kind (#68)

obligation-check.sh: a review file with dated addenda FAULTed because every
proof line was compared with the single newest stamp. A citation now belongs
to the nearest stamp above it (headline's if above all). _ob_stale picks the
stamp with the newest commit time, not the last one in the file.

brd-to-build-plan.md: rationale for not adding a LOOK kind next to the Kind list.

Fixture: tests/wave2/test-obligation-valid-at.sh T4-T8 (3 fail on master).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
process: review-rule fixes from its own independent review
7168174 
Reviewer inputs stated (diff, issues, the PR body what-section, never its verdicts);
durable Review: line in the squash commit; unblock-master exception per guard
rule 7; budget stated as the render-routing ratchet, 95% of it; verified figures.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
journeys: #302 follow-up — accept a project-set password, SIGINT find…
cc56d9b 
…ings, setup kind

- config.js reports passSource (env / project.config.js / template
  placeholder); the preflight refuses only the template placeholder
  (Demo2025!, or credentials() source 'template default'). As shipped,
  #302 refused every non-env password, though the template tells projects
  to put their real one in DEFAULT_PASS and nothing loads an env file
  before verify-module — every journeys rung would FAULT after a sync.
  An admin object without passSource keeps the #302 rule.
- SIGINT/SIGTERM write the findings slot as aborted ("signal SIGINT")
  instead of leaving the previous run's file there.
- report-normalize.js: rung 'setup' rows keep kind 'setup' (were 'ui').
- CHANGELOG names ALLOW_CONFIG_CREDENTIALS and the 403 verdict change
  (data rungs INVALID exit 1 -> suite aborts exit 2).

test-journey-preflight.sh 74/74 (9 new cases fail on master: project-set
password runs, placeholder refuses, env wins, SIGINT end to end);
report-normalize --selftest new setup-kind case.

Refs #286

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
fix(lint-gate): a Starlark rule installed after the baseline is FIRST…
5432345 
…-SEEN, not a rise (#305 review)

A rule sync-project.sh installs after the baseline has no baseline count, so its first hit
compared against 0: every exec logged LINT ROSE and exited 1 until someone re-baselined, and
the Lint Batch 3 rules fire on ordinary text (TBD, XXX, v1.2, "12 Oct 2026").

- project-bin/lint-gate.sh: a rule absent from the baseline counts is FIRST-SEEN when it is a
  .star rule in .claude/lint-rules/ (RULE_ID = "...") the baseline did not know: not in the new
  `star_rules` list --update-baseline now writes, or, for an older baseline, its .star file is
  newer than the baseline. Printed as `FIRST-SEEN <rule> N finding(s) — new rule, first seen:
  review and re-baseline with --reason`; exit 0 when nothing else rose; verdict
  PASS-FIRST-SEEN. Re-baselining it needs --reason (it raises a count), after which it is an
  ordinary ratcheted rule. Built-in mxcli rules and .star rules known at baseline with zero
  findings still count as a rise (a baseline omits zero-count rules).
- project-bin/exec.sh: the rc-0 row carries `FIRST-SEEN: <rule> (N) — review, re-baseline with
  --reason`.
- tests/wave2/test-lint-gate-first-seen.sh: real toolkit .star files, LINT_JSON input; 17/17
  here, 8 fail on master's lint-gate.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
final-look: PROOF-OF-LOOK + VALID AT required; project/cluster rows N…
8c0b176 
…OT YET DUE before their stage

From an independent review of #303. obligation-check.sh compared from-stage
only with --adopt, so final-look (stage 6) read PENDING from Stage P. A
project/cluster row above the register's Current stage now reads NOT YET DUE
(module rows unchanged; an unreadable stage line keeps the old owed verdict).
The PROOF-OF-LOOK check now covers final-look as module-review.md requires,
and an unstamped final-look report is FAULT (it could never go stale).

Fixture: tests/wave2/test-obligation-final-look.sh T4-T10 (4 fail on master).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
page-fidelity: match text on what the page shows; ButtonStyle classes…
944fce9 
…; chrome from layout (#158, #157)

#158: headings/actions/content were a substring test over the whole page MDL
(page name, identifiers, comments) with 60% of the >3-char words enough. Each
wireframe string is now matched per literal against the page's own captions
of the right kind, on whole words, all words for labels under three words;
a label with only short words is required instead of passing vacuously.
Bindings keep the identifier corpus.

#157 remainder: an actionbutton declares btn + btn-<ButtonStyle> (btn-default
when DESCRIBE omits it); the page's layout MDL in the input, or --chrome a,b,
adds to the fixed chrome set. Without either, nothing changes.

Before/after over every page/wireframe pair in the repo: 11 of 13 unchanged;
the --stub commented-template page 66% -> 39% (a heading credited from
dgWaiting + CustomerQuery identifiers); two new probe pairs fail on master.
Fixtures: test-page-fidelity-mocks.sh +16 assertions (13 fail on master).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
Merge remote-tracking branch 'origin/chore/baseline-word-trim' into f…
aaa9890 
…eat/interaction-model-skill
@claude
fix(exec): follow-ups from independent review of #304/#306/#290
ed4efbb 
- Stale-lock takeover serialised through an O_EXCL .exec.lock.takeover file;
  the lock is removed only if it still holds the stale line read (no mv -f
  restore). 12 contenders x 10 rounds: master 2/10 rounds with 2-3 holders,
  fixed 0/10.
- --watch-compat: only ui-loop.md's watch-safe list says "watch can apply
  this"; other statements print "unknown — restart if behaviour looks stale".
- Two-tree (.mpr under app/) cases for the lock and --commit-first.
- Leading flags parsed in one loop (--commit-first --cr CR-NN works); a
  script name starting with -- is refused before any commit or snapshot.
- ui-only test lane names the embedders of a changed snippet/layout.

Refs #287 #289 #274

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
chore(trim): correct the measured word counts in the CHANGELOG line
cd67a7c 
Measured with LC_ALL=C on the branch after master (#313) was merged in:
79,997 -> 76,259 (-3,738); the old line used pre-#313 figures.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
@claude
process: second review round on the merge-review rule
234e06d 
Testing section gets the reviewer fixture exception it was cited for; the
PENDING verdict goes to a PR comment (squash bodies are immutable); the trim
trigger is within 2,000 words of the budget after the PR, not 95%.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
This was referenced Oct 14, 2026
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
Closed
MendixMau pushed a commit that referenced this pull request Oct 14, 2026
@claude
docs: review fixes for #334 (tags start at 1.0, pinning, PR links, ca…
f279645 
…dence pointer)

- v1.0.md: 1.0 is the first published train tag; note on PRs closed via #329;
  sync prints "none tagged yet" until tags exist; app-restart.sh is #304,
  #322 added the injection refusal; "UX001/002 plus Batches 1-4"; known gap
  for the stale-lock takeover race after a killed holder.
- RELEASE-PROCESS.md §8: tags start at 1.0; placeholders instead of a concrete
  tag; session ritual warns while pinned (expected); cadence defers to the
  CHANGELOG header instead of restating it.
- README.md: Toolkit release line before/after tags; tags start at 1.0;
  interview-mode --set persists in .claude/.interview-mode, env var is
  per-session; "alias: unattended" next to auto.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VJgWP5vEoAsNsJYqCDMGNw
This was referenced Oct 14, 2026
Merged
Closed
@MendixMau
MendixMau deleted the integration/2026-10-14 branch October 15, 2026 10:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@MendixMau @claude