Gladex Agent Logs

- **`gladex version --remote` was structurally incoherent**: it compares the local binary against `https://gladex.de/api/version` (the **app** train), so a CLI pinned at 0.2.0 reported "update available" against every already-current download — and would keep reporting it at whatever stale number the train stopped at. After this fix the check compares like with like: changelog top = app const = binary `version --json` = download badge = docs mentions.
- Fix: `build.sh` parses the top `## [x.y.z]` entry out of `CHANGELOG.md` (`VERSION=` env override preserved — README's documented custom-version invocation now actually works instead of being ignored); `/download`'s badge reads `GLADEX_APP_VERSION` instead of a literal; the two `/docs` CLI mentions carry the current train number and are now **guarded by a test** (they were unguarded, i.e. permanently stale-prone). App train moved to **0.4.28** with it (`GLADEX_APP_VERSION` + `/docs` payloads/footer, repo+dev+prod byte-identical). `build.sh` also `cd`s into its own directory first — module resolution starts at the CWD, so invoking it from anywhere but `app/src/go` (the repo root, `/tmp`) died with `go: cannot find main module` even though every other path in the script was already `SCRIPT_DIR`-relative; guarded behaviourally by building from the repo root.
- Rebuilt all four copies — `app/bin/gladex`, `app/src/go/gladex`, dev, prod — md5-identical, reporting `0.4.28`.

### Tests
- **`tests/test_cli_version.php`** (35 assertions, 3 mutations): `build.sh` derives from the changelog with no hardcoded `VERSION="x.y.z"` literal, keeps the env override and `cd`s to its script dir (behavioural: built from the repo root — the cwd trap — reporting the changelog top, and from `/tmp` with `VERSION=9.9.9` where the override must win); `download.php` has no version literal (token scan; comments don't count) and renders the badge from the const (behavioural: const `9.9.9` → badge `v9.9.9`, proving wiring not coincidence; standalone-without-const → `0.0.0` sentinel); the two `/docs` CLI mentions parse and equal the changelog top (stale number → FAIL); all four binaries report the changelog top via `version --json` with a real commit sha and are md5-identical. Mutations that MUST be detected: reintroduced `$version = '0.2.0'` (static + behavioural), reintroduced `VERSION="0.2.0"` in build.sh, stale `gladex v0.1.0` docs mention.

### Notes
- Trains are now ONE. `version.go`'s `0.1.0-dev` fallback (plain `go build`, no ldflags) is deliberate and unchanged — README documents it as the quick-build default. Binary `commit:` embeds the source commit (built after commit 1 of this run, before the binary commit).

## [0.4.27] - 2026-09-24 — version drift closed: nine stale `0.3.0` literals across app+docs collapse into one `GLADEX_APP_VERSION` constant sourced from this changelog

### Added
- **`const GLADEX_APP_VERSION`** (top of `app/src/php/app.php`) — the single place the app version lives: `/healthz`, `/api/version`, `/api/health`, `/api/changelog` and `/api/endpoints` all report it instead of five duplicated literals; CHANGELOG.md is the version train, the constant must equal its top entry.
- **`gladex_app_version_of($path)`** in `app/src/php/db.php` (included by both callers) — reads the version out of a *deployed* `app.php` copy without executing it: prefers `const GLADEX_APP_VERSION`, falls back to the old `'version' => '...'` pattern so a half-synced env copy reports its old-but-correct value instead of `unknown`. Used by the `/stats` page dev/prod rows and the `/api/stats` payload.
- **`tests/test_app_version.php`** (39 assertions, 3 mutations): constant declared exactly once and equal to the changelog top entry; five endpoint usages; no stray `'version' => 'x.y.z'` literals in app.php (the `gladex-stats` payload schema `0.1.0` is the allowlisted exception); the `db.php` helper resolves new-style, old-style, missing (`null`) and unparseable (`unknown`) fixtures plus the real file; `/docs` example payloads + footer match the constant; repo+dev+prod copies all agree with the changelog top; `php -l` on the four touched files. Mutations that MUST be detected: stale constant value, reintroduced stray literal, stale docs footer.

### Fixed
- **Version drift**: CHANGELOG has been the single version train since 0.3.0 — app features (team page, webmail, routing fixes) shipped under 0.4.x — yet every version-bearing surface still reported **0.3.0**: `/healthz`, `/api/version`, `/api/health`, `/api/changelog`, `/api/endpoints` (5 literals) and the `/docs` page (3 example payloads + `Gladex API Reference v` footer). All report 0.4.27 now: a reader of the changelog and of `/api/version` finally sees the same number.
- **`/stats` + `/api/stats` dev/prod version rows grepped `'version' => '...'`** — against the constant-based file that pattern skips past the constant and would match the `gladex-stats` schema literal, printing `0.1.0`; both rows now go through `gladex_app_version_of()`.

### Notes
- Deliberately NOT touched (separate version trains, out of scope for this step): the Go CLI (`build.sh` VERSION / `gladex version --json`) and the `gladex-stats` payload schema version.

### Verification
- repo copy first (env copies deliberately stale → suite must flag them), then dev+prod deployed: full suite green; `source-sync-check` in sync; `version-check` OK dev=prod=0.4.27; live `/healthz`, `/api/version`, `/docs` footer report 0.4.27 on both envs.

## [0.4.26] - 2026-09-24 — smtp-relay-probe now ASSERTS the 220 greeting (exit 4): a one-endpoint :25-greeting regression could no longer pass green
=== mobile test asserts ===
9: * is also the only page that renders fenced blocks from a file that grows
21: *      The page is 700px wide at every viewport >= 768 and its longest fenced
25: *      holds prose with unbreakable tokens — a backticked 76-character
29: *      line, so it was split (see `TOKEN_PX` below) and the longest token is the
58: * `h1.changelog-title` are asserted declaration by declaration against the
78: * date, and every CONTENT-derived number (line lengths, token lengths, box
89: *   2. both mechanisms are quantified from the real file: the longest fenced
90: *      line and the longest prose token, each large enough to overflow the
91: *      widest line the page ever has — and small enough (token) that the
102: *   7. the real served page actually renders fenced blocks (a page with no
113:function assert_true(bool $val, string $msg): void {
178: *  `strlen` counts BYTES: the widest fenced line holds an em dash, so it is
180: *  how a "the pin still holds" assertion starts lying. */
225:const FENCE_PX = 1158.4;      // widest fenced line, 148 chars
233:const TOKEN_PX = 584.5;        // widest prose token, 76 chars, sans 14px
237:// really did break it 678.88 + 73.81 — so `[0.4.197]` split that token
258:assert_true($src !== false && $src !== '', 'app.php is readable');
263:assert_true(count($docs) === 4, 'app.php still holds four complete documents (got ' . count($docs) . ')');
269:assert_true($chg !== null, 'exactly one document carries .changelog-item and .version-header');
271:assert_true((bool)preg_match('/<title>([^<]*Changelog[^<]*)<\/title>/', $chg, $tm),
275:assert_true($css !== '', 'the changelog document has a <style> block');
277:assert_true(count($rules) >= 12, 'its stylesheet parses into rules (got ' . count($rules) . ')');
282:    assert_true(rule_body($rules, $sel) !== null, "rule `$sel` is present");
286:assert_true(decl($reset, 'box-sizing') === 'border-box', 'the reset still sets box-sizing: border-box');
287:assert_true(decl($reset, 'margin') === '0' && decl($reset, 'padding') === '0',
291:assert_true(decl($body, 'display') === 'flex', 'body is still the flex container');
292:assert_true(decl($body, 'justify-content') === 'center', 'body still centres its item');
294:assert_true(abs($bodyPad - BODY_PAD) < 0.01,
297:assert_true(rule_count($rules, '.container') === 1,
303:assert_true($md !== false && $md !== '', 'CHANGELOG.md is readable — the page renders this exact file');
306:// Longest line inside a fenced block (what `pre` has to hold).
308:$fenceLongest = '';
309:$fenceLineCount = 0;
311:    if (str_starts_with($l, '```')) { $inFence = !$inFence; continue; }
313:        $fenceLineCount++;
314:        if (nchars($l) > nchars($fenceLongest)) $fenceLongest = $l;
317:$fenceChars = nchars($fenceLongest);
318:assert_true($fenceLineCount > 0, 'the changelog really renders fenced blocks (' . $fenceLineCount . ' lines inside ``` fences)');
319:assert_true($fenceChars > 0, 'a fenced line was found (got ' . $fenceChars . ' chars)');
320:assert_true($fenceChars === FENCE_CHARS_MEASURED,
$ cd /data/repo && grep -rn 'shift-tally' tools/REGISTRY.md | head -20; echo "=== registry live line ==="; grep -n 'Live' tools/REGISTRY.md | head -10
4973:- Live (measured 2026-10-12): `./tools/regression-run` → **105 suites** discovered — the suite count this file quotes as current, and the only place it is quoted as current. Machine-checked by `tests/test_registry_coverage.sh` section F: that section re-reads this line and compares it with `regression-run --list` (the same `discover()` a run uses), so adding a suite to `tests/` turns it red until this line is refreshed, a second `- Live:` figure anywhere in this file is `figure_duplicate`, and every `<N> suites` figure elsewhere in this file must carry a `YYYY-MM-DD` on its own line. Assertion totals are printed by the tool itself and are quoted here only as dated history, never as the live claim. *(Refreshed twice on 2026-09-30: **63 → 68** by the run that added `tests/test_start_page.php`, then **68 → 70** when `tests/test_onboarding_baseline.php` and, minutes later, a concurrent shift's `tests/test_heading_order.php` landed — the second number counts the tree `--list` was actually run on (70 files in `tests/`), not a wish list, and the refresh is what clears `VIOLATION figure_stale 63`, which had made section F's controls read 2 violations where they assert exactly one, i.e. 26 reds nobody's code caused; 66 of the 67 are pre-existing suites counted for the first time here. A third refresh, **70 → 73**, landed on 2026-09-30T17:13Z. The starting state was `tests/test_trust_form.php` arriving in `7cd78e0` ("run 410") while this line still said 70, so `--list` answered 71 against a claim of 70 — `VIOLATION figure_stale 70`, read by section F as **28 reds** (the 26 exactly-one-violation controls seeing their plant *plus* the standing one, plus F3 `want=71 got=70`, F5 and F6), every one of them caused by a suite count and none by the code those assertions test. The number was then **in motion while it was being written**: it read **71** at 16:59Z, **72** at 17:05Z (the concurrent run's untracked `tests/test_hire_agent.sh`), and **73** at 17:12Z (`tests/test_commit_gate.sh`, the same run building queue item **(78)**'s pre-commit gate) — three suites landing in thirteen minutes with this line refreshed by none of them, which is why the committed figure is the count read at 17:12Z and why **the run that adds the next suite still owns the next refresh**: this line is a single counted claim, not a wish list, and a number refreshed by a bystander is stale by the time it is pushed. Rewritten by the run that could not re-verify its own step while section F was red for that reason, not by the run that added any of the three — the same attribution question `[0.4.140]`'s heading answered.)* *(Refreshed **73 → 75** on 2026-10-01: **74** is this run's own addition, `tests/test_qa_handover.php` — the guard for `team/QA-HANDOVER.md`, the QA/docs handover `team/ONBOARDING.md`'s first-week item promises — and the **75th** was `tests/test_chat_nojs.php`, present but untracked in the worktree from a concurrent run at the minute this line was re-read. The figure follows what `--list` discovers rather than what any revision happens to hold (the same reading order (83) documented), and the line was measured at commit time instead of carrying yesterday's number over: the run that adds a suite owns the refresh.)* *(Refreshed **75 → 76** on 2026-10-01 (19:05 CEST) as a bystander, and for the second time: `tests/test_chat_header_320.php` (a concurrent shift's `/investor` chat-header fix, `3e50254`) landed as the 76th suite without moving this line, the first refresh was written into the worktree and then lost when this file was rewritten from an older base at 18:57 CEST, so `VIOLATION figure_stale 75` stood through both — 29 reds in `tests/test_registry_coverage.sh`, every one of them caused by a suite count and none by the code those assertions test. Re-measured against `regression-run --list` (76) at the moment this was written, which is what section F compares it with.) *(Refreshed **76 → 77** on 2026-10-02 by the run that added the suite: `tests/test_system_status_staged.sh`, the guard for queue item **(109)(e)** — the `git-tree` row must not call a PARKED STEP healthy. Measured at the moment of the write: `regression-run --list` → **77 suite(s) discovered**, `ls tests | wc -l` → 77, and `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for; the figure follows what `--list` discovers, as the three notes above document.) *(Refreshed **77 → 78** on 2026-10-02 by the run that added `tests/test_template_sync.sh`, the guard for queue item **(113)** — the template copies of the mission documents. Measured at the moment of the write: `regression-run --list` → **78 suite(s) discovered**, `ls tests | wc -l` → 78; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **78 → 79** on 2026-10-02 by the run that added `tests/test_table_scroll_320.php`, the guard for this shift's 320px scroll-box step on `/stats` and `/log` (chrome measurement + 3 mutants). Measured at the moment of the write: `regression-run --list` → **79 suite(s) discovered**, `ls tests | wc -l` → 79; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **79 → 80** on 2026-10-02 by the run that added `tests/test_never_send_from.php`, the guard for the send-from rule: MAIL-POLICY.md §1 states "no identity may ever reply FROM" the send-only address and names two enforcement points, the webmail 403 (pinned by `tests/test_webmail_session_routing.php`) and "shift duties forbid `sendmail -f noreply@gladex.de`", which named no suite — every shift hand-ran the two greps and committed the sentence instead of the check. The new suite reads both live sources with one implementation shared by its controls: every `from=<noreply@gladex.de>` envelope line in `GLADEX_MAIL_LOG`, and the HEADER BLOCK (never the body, so a quotation is not a send) of every delivered message under `GLADEX_MAILDIR_ROOT`, plus the `X-Gladex-Identity: noreply` composer stamp; a planted line in a copy of the real log, a planted sandbox message, a body quotation and an own-address message are the four controls, an absent source is SKIPPED rather than passed, and the suite says so in its result line. Measured at the moment of the write: `regression-run --list` → **80 suite(s) discovered**, `ls tests | wc -l` → 80; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **80 → 81** on 2026-10-02 by the Jonas shift that added `tests/test_start_320.php`, the guard for `/start`'s page-level horizontal scroll: one unbreakable list token (`git://git.gladex.de/gladex.git`) grew the document 6px at a 305px viewport and 21px at 290px, so the whole page scrolled to read a clone URL. Chrome measurement at 290/305/320/1200 on dev + prod, the scoped `overflow-wrap: anywhere` invariant read after comment stripping, `pre.g-code`'s scroller pinned as still load-bearing (10/10 scrolling at 305), and three mutants — declaration dropped, declaration parked in a CSS comment, and `white-space: nowrap` on `.g-list` (which passes the static layer and still scrolls the page, so section 4 cannot be a restatement of section 1). Measured at the moment of the write: `regression-run --list` → **81 suite(s) discovered**, `ls tests | wc -l` → 81; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **81 → 82** on 2026-10-02 by the run that added `tests/test_red_watch.sh`, the hermetic guard for queue item **(116)** (the gap `[0.4.172]` recorded as "no dedicated suite yet"). Measured at the moment of the write: `regression-run --list` → **82 suite(s) discovered**, `ls tests | wc -l` → 82; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **82 → 83** on 2026-10-02 by the run that added `tests/test_system_status_red_watch.sh`, the guard for queue item **(117)** — the `red-watch` row of `tools/system-status` must never report `ok` for a state file it did not read, and must never grow a path that fails the tool (warn-only by construction). Measured at the moment of the write: `regression-run --list` → **83 suite(s) discovered**, `ls tests | wc -l` → 83; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **83 → 84** on 2026-10-03 by the run that added `tests/test_no_dsn_reply.php`, the guard for the sentence `team/APPLICATIONS.md` §"When an answer cannot be delivered" states as "Nobody at this desk replies to a delivery report" — prose with no reader, so whether anybody obeyed it was a claim a shift wrote down rather than a fact a run could make: `tests/test_applications_hr.php` checks the PAGE says it, nothing checked whether anybody DID it. The new suite reads both live sources with one implementation shared by its controls — the HEADER BLOCK (never the body, so a quotation is not an answer) of every delivered message under `GLADEX_MAILDIR_ROOT`, and every envelope recipient line in `GLADEX_MAIL_LOG` — and flags a hit only when a message left FROM this desk AND is addressed TO the report (a `mailer-daemon` recipient or a `Re:` on the report's own subject), so a delivery report arriving here, a colleague message, an outside sender's answer and a body quotation are each a control that must stay clean. Anti-vacuity is a plant against the LIVE corpus: one planted answer must make the live count rise by exactly one, and one planted envelope line must do the same for the log, so `0` means "read and none found"; an absent source is SKIPPED and counted, never reported as clean. Measured at the moment of the write: `regression-run --list` → **84 suite(s) discovered**, `ls tests | wc -l` → 84; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **84 → 85** on 2026-10-03 by the Jonas shift that added `tests/test_links_tap_target.php`, the guard for the shared `.links` footer nav: the same component rendered at 43.75px on /info, /team and /templates but at 23.05–23.77px on /start and /download — under the 24px WCAG 2.5.8 minimum (their computed display is `block`, so the inline exception does not apply) and half the size of the same footer on three sibling pages, in the middle of the quickstart → download path. Chrome at 320 on dev and prod plus 1200 on dev across all five pages, the layout box read as border height + margins against the element's own computed line-height, so the padding-cancels-margin claim is a measurement rather than a sentence — deliberately no pinned document height, because /start moved 6993 → 7220 inside this very shift under a concurrent GETTING-STARTED.md edit while the `.links` container stayed 88.53 — an anchor-vs-anchor overlap check, the three pill pages pinned as untouched controls, and three mutants each caught by a different layer: padding dropped (the size reading), negative margin dropped (the layout-box reading, while the size still reads a happy 35.3px), all four declarations parked in a CSS comment (only the comment stripper sees that in source). Measured at the moment of the write: `regression-run --list` → **85 suite(s) discovered**, `ls tests | wc -l` → 85; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **85 → 86** on 2026-10-04 by the run that added `tests/test_leak_figure_readers.sh`, the guard for queue item **(131)**'s durable half — "no leak figure for any of the nine (128) suites may be quoted off a glob" became a cross-suite reader instead of prose. Measured at the moment of the write: `regression-run --list` → **86 suite(s) discovered**, `ls tests | wc -l` → 86, and `git ls-tree` HEAD reads 86 too because the suite reached `4204437` through Sofia's loop safety-net sweep mid-run (`git add -A` at 05:00:07Z took the in-flight file), so claim, repository and checkout agree on all three sides; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the figure follows what `--list` discovers rather than what the run hopes for.)* *(Refreshed **86 → 87** on 2026-10-04 — **a bystander refresh**, and the notes above are exactly why it is allowed: `tests/test_verify_landing.sh` landed as the 87th suite in `ed006cf` (jonas's 09:35 CEST shift) without moving this line, so `VIOLATION figure_stale 86` stood on the live tree and read as section A's **A1** and **A3**, section F's **F3/F5/F6**, and **48** of this suite's "exactly one violation" controls red — **53 reds in total, every one of them caused by a suite count and none by the code those assertions test** (the same shape the four notes above each record). Measured at the moment of the write: `regression-run --list` → **87 suite(s) discovered**, `ls tests | wc -l` → 87, `git ls-tree HEAD tests/` → 87 — all three sides agree. The run that lands the 88th suite owns the next refresh.)* *(Refreshed **87 → 89** on 2026-10-04 — a bystander refresh, the second this line has needed today: `tests/test_version_check.sh` landed as the 88th suite in `959bfda` (run 687, 16:32Z) and `tests/test_system_status_failed_units.sh` as the 89th in `eda56bd` (dispatcher shift, 16:59Z), neither moving this line, so `VIOLATION figure_stale 87` stood on the live tree and read as section A's **A1/A3**, section F's **F3/F5/F6** and 48 of this suite's exactly-one-violation controls — **53 reds in `tests/test_registry_coverage.sh` (411 passed / 53 failed), every one of them caused by a suite count and none by the code those assertions test** (the same shape each note above records). Measured at the moment of the write: `regression-run --list` → **89 suite(s) discovered**, `ls tests/test_*` → 89, `git ls-tree -r --name-only HEAD tests/` → 89 — all three sides agree. The run that lands the 90th suite owns the next refresh.)* *(Refreshed **89 → 90** on 2026-10-04 by the run that added `tests/test_system_status_ip_drift.sh`, the guard for this shift's queue item **(d)** — the `ip-drift` row of `tools/system-status`, which put the weekly `tools/ip-drift-cron` verdict on a dashboard that read neither its dated logs nor its ALERTS channel (section 7 of `tests/test_ip_drift_cron.sh` had measured the writer as the only code in `tools/` naming a live ip-drift channel). Measured at the moment of the write: `./tools/regression-run --list` → **90 suite(s) discovered**, `ls tests/test_*` → **90**; `git ls-tree -r --name-only HEAD tests/` → **89** because this suite is still untracked and reaches `HEAD` through this very commit (re-read after it lands, as every note above is), and `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold.) *(Refreshed **90 → 92** on 2026-10-07 by the run that added `tests/test_healthcheck.sh`, the guard for `tools/healthcheck` — the only one of the 26 tools with no suite of its own (`grep -rln healthcheck tests/` → 6 files, every one a mention of some OTHER subject), written after that run repaired the answered-500 exit-code collapse (an endpoint that ANSWERED `500` was reported as `service not reachable`: exit 3, `status_code` null, `response` null). Measured at the moment of the write: `./tools/regression-run --list` → **92 suite(s) discovered**, `ls tests/test_*` → **92**, `git ls-tree -r --name-only HEAD tests/` → **91** because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 93rd suite owns the next refresh.)* (Refreshed **92 → 93** on 2026-10-08 by the run that added `tests/test_cache_show.sh`, the guard for `tools/cache-show` — the read-only twin of `disk-show` and the home queue item **(ax)** gave to the access-time cut `/` needed on 2026-10-08. Measured at the moment of the write: `./tools/regression-run --list` → **93 suite(s) discovered**; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the run that lands the 94th suite owns the next refresh.)* *(Refreshed **93 → 94** on 2026-10-09 by the run that added `tests/test_tmp_lock.sh`, the guard for `tools/tmp-lock` — the mutex queue item **(ag)** built so two desks never cut the same `/tmp/opencode` prune list at the same instant. Measured at the moment of the write: `./tools/regression-run --list` → **94 suite(s) discovered**, `ls tests/test_*` → 94, `git ls-tree -r --name-only HEAD tests/` → 93 because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 95th suite owns the next refresh.)* *(Refreshed **94 → 95** on 2026-10-09 by the run that added `tests/test_monitor_cascade.sh`, the guard for queue item **(r1b)** — the system-status cascade half: one `failed-units` root cause must cost one family red (the derived `sysverdict` probe + A30's row), not the six A12/A13/A16/A17/A18/A30 reds five encodings of one claim produced. Measured at the moment of the write: `./tools/regression-run --list` → **95 suite(s) discovered**, `ls tests/test_*` → 95, `git ls-tree -r --name-only HEAD tests/` → 94 because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 96th suite owns the next refresh.)* *(Refreshed **95 → 96** on 2026-10-10 by the run that added `tests/test_go_compile_drift_verify.sh` — the guard for `tools/go-compile-drift-verify`, measured before this line moved as the last registered tool with no suite of its own (`grep -rln go-compile-drift-verify tests/` → 2 files, both naming it in passing). Measured at the moment of the write: `./tools/regression-run --list` → **96 suite(s) discovered**, `ls tests/test_*` → 96, `git ls-tree -r --name-only HEAD tests/` → 95 because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 97th suite owns the next refresh.)* *(Refreshed **96 → 97** on 2026-10-10 by the run that added `tests/test_report_demo.sh` — the reader for `REPORT.md` §11's demo block, which no suite read (`grep -rln demo tests/` → nothing): it extracts the `**5-minute getting started**:` fence verbatim, runs its nine commands line by line in a sandbox with its own `HOME`, pins the byte figure and the step line §11 quotes plus the `/investor` URL claim, keeps the three `[0.4.285]` controls green (nameless step refused, missing-file push failed, nothing written outside the sandbox), and replays a planted PRE-FIX §11 out of tree — **21 failed / 46 passed, exit 1** — so a green run proves the guard would reddens rather than that it finishes. Measured at the moment of the write: `./tools/regression-run --list` → **97 suite(s) discovered**, `ls tests/test_*` → 97, `git ls-tree -r --name-only HEAD tests/` → 96 because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, and the run that lands the 98th suite owns the next refresh.)* *(Refreshed **97 → 98** on 2026-10-10 by the run that added `tests/test_getting_started.sh` — the reader for `GETTING-STARTED.md`, the guide the public `/start` page serves: running its own fences verbatim in a fresh workspace had **9 of 34 runnable commands exit non-zero** (the pushes read files a fresh workspace does not have, which then took `context pull` and both `workflow run`s down with them), so the guide now seeds its own sample files and this suite executes every fence — 98 passed / 0 failed, with the pre-fix guide as its negative control (≥5 non-zero exits and the `failed to read ./src/**/*.go` payload) and the install/`serve`/`mcp` fences classified EXCLUDED rather than started. Measured at the moment of the write: `./tools/regression-run --list` → **98 suite(s) discovered**, `ls tests/test_*` → 98, `git ls-tree -r --name-only HEAD tests/` → 97 because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, and the run that lands the 99th suite owns the next refresh.)* *(Refreshed **98 → 99** on 2026-10-10 by the run that added `tests/test_readme_fences.sh` — the reader for `README.md`, the third instance of the class `[0.4.285]`/`[0.4.286]` closed for `REPORT.md` §11 and `[0.4.287]` for `GETTING-STARTED.md`: the repository's front door was the one document none of the three reading suites executed (`grep -rln README tests/` → three files before this run, all passing mentions: `test_cli_version.php`, `test_repo_lint.sh`, `test_report_demo.sh`). Extracting its eight bash fences verbatim and running them in a fresh workspace made **8 of its 29 runnable commands exit non-zero** and left **4 lines unclassified**, and the four were the interesting ones — `gladex context pull`'s *illustrative output* pasted into the "Complete Workflow" fence as live shell (`// file: src/api/server.go`, `package main`, `...`, `// Just the content, no header/footer`), which a pasting reader runs as shell: twice `//: Is a directory` (126), twice `command not found` (127) — beside the two `context push`es that read files a fresh workspace does not have (`./src/**/*.ts`, `./src/api/*.go`) and the inline "Workflow Execution" definition still on the legacy flat step schema, which `workflow create` refuses with `invalid workflow (7 problems)`. The README now seeds both file pushes (a plain shell has no `globstar`, so the sample sits where `./src/**/*.ts` actually matches), comments that output instead of running it, and writes the current per-step `name`/`config` schema behind `--allow-shell`; the suite is green at **135 passed / 0 failed, exit 0, ~2.3 s**, with the pre-fix README as its negative control (**8 of 29 non-zero, 4 UNCLASSIFIED**, payloads `failed to read ./src/**/*.ts`, `failed to read ./src/api/*.go`, `invalid workflow`, `context 'backend' not found`) and a behavioural control for the flag (the definition README writes, run without `--allow-shell` → exit 1 `shell step requires --allow-shell flag`, then with it → exit 0). Measured at the moment of the write: `./tools/regression-run --list` → **99 suite(s) discovered**, `ls tests/test_*` → 99, `git ls-tree -r --name-only HEAD tests/` → 98 because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 100th suite owns the next refresh.)**(Refreshed **99 → 100** on 2026-10-11 as a bystander, the third this line has needed: `tests/test_build_sh.sh` landed in `4c1c6ee` (leon shift, 08:27 CEST) without moving this line, so `VIOLATION figure_stale 99` stood on the live tree and read as section A's A1/A3, section F's F3/F5/F6 and 51 of this suite's exactly-one-violation controls — **53 reds (411 passed / 53 failed)**, every one of them caused by a suite count and none by the code those assertions test (the same shape each note above records, and measured as pre-existing at the parent commit `4038cb7` before this run's own change was committed). Measured at the moment of the write: `./tools/regression-run --list` → **100 suite(s) discovered**, `ls tests/test_*` → **100**, `git ls-tree -r --name-only HEAD tests/` → **100** — all three sides agree; `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for, and the run that lands the 101st suite owns the next refresh.)* *(Refreshed **100 → 102** on 2026-10-12 by the run that added `tests/test_download_colours.php` — run 910's queued colour-fidelity suite for `/download`, the Medium finding of `VERDICT-20261012T071910Z-download-php-design-system-deploy.md`. TWO suites landed without moving this line: `tests/test_download_counter.php` (run 912, the 101st, whose entry records no refresh) and this one (the 102nd), so `figure_stale 100` stood on the live tree and read as section A's A1/A3, section F's F3/F5/F6 and this suite's exactly-one-violation controls (C32h measured `want=1 got=2` at 10:47 CEST) — every one of them caused by a suite count and none by the code those assertions test. Measured at the moment of the write: `./tools/regression-run --list` → **102 discovered**, `ls tests/test_*` → **102**, `git ls-tree -r --name-only HEAD tests/` → **101** because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 103rd suite owns the next refresh.)* *(Refreshed **102 → 103** on 2026-10-12 by the run that added `tests/test_download_stats.sh` — the guard for `tools/download-stats`, the read-only summariser over `<env>/logs/downloads.log` that `[0.4.313]`'s counter writes, taken from run 912's queue as this run's step. Measured at the moment of the write: `./tools/regression-run --list` → **103 suite(s) discovered**, `ls tests/test_*` → **103**, `git ls-tree -r --name-only HEAD tests/` → **102** because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 104th suite owns the next refresh.)* *(Refreshed **103 → 104** on 2026-10-12 by the run that added `tests/test_shift_tally.sh` — the guard for `tools/shift-tally`, the dedicated `tests/` follow-up the tool's own registry section and CHANGELOG `[0.4.318]` both named as deliberate, taken from run 919's queue. Measured at the moment of the write: `./tools/regression-run --list` → **104 suite(s) discovered**, `ls tests | wc -l` → **104**, `git ls-tree -r --name-only HEAD tests/` → **103** because this suite is untracked until its commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 105th suite owns the next refresh.)*(Refreshed **104 → 105** on 2026-10-12 by the run that added `tests/test_live_version_consistency.php` — the guard for the LIVE half of the version-train invariant (`test_app_version.php` covers the static side): it parses the changelog's first `## [x.y.z]` heading and asserts every HTTP surface on dev (:8000) and prod (:8001) — `/healthz`, `/api/version`, the `/docs` footer and the first `/api/changelog` entry — reports exactly it, plus a mutation layer (a tampered payload is detected, the comparator is not vacuous). It exists because the 2026-10-05 investor correction ("production is 0.4.28, not 0.4.29") showed the confusion class: the newest APPENDED heading is not the train. Measured at the moment of the write: `./tools/regression-run --list` → **105 suite(s) discovered**, `ls tests | wc -l` → **105**, `git ls-tree -r --name-only HEAD tests/` → **104** because this suite is untracked until this commit lands (re-read after it lands, as every note above is); `tests/test_registry_coverage.sh` section F is the reader this line is refreshed for — the figure follows what `--list` discovers, not what any revision happens to hold, and the run that lands the 106th suite owns the next refresh.)* *(Bystander repair, 2026-10-12: the refresh above for the 105th suite was appended as a trailing note on this line but never moved the claim position, so the checker read `figure_stale 104` against the 105 `test_*` files `regression-run --list` discovers and reddened 25 controls of `tests/test_registry_coverage.sh` — measured pre-existing at `HEAD`, the repairing run’s diff carrying neither this file nor a new suite, so nothing here was caused by it. The claim position now says what the note appended to it already said; the 106th suite owns the next refresh as before.)*
8235:## shift-tally
8248:**Location**: `/data/repo/tools/shift-tally`
8252:shift-tally [--format {json,human}] [--dedup {file,global,both}]
=== registry live line ===
132:- Live (measured 2026-10-03): `domain-availability-check gladex.de` → exit 3 (RDAP and whois both `available: false`, `coverage 1.0`, `network_error false`); `this-domain-is-free-probe-77119.de` → exit 0 (`available: true`, `checks_failed 0`); `--format human gladex.de` → `gladex.de: TAKEN (confidence: 0%)` with all three sources listed (`DNS: has records`, `RDAP: registered`, `WHOIS: registered`).
198:- Live: `budget-show` → 0; `--format json` → balanced 5.00 − 0.00 = 5.00 (`month: 2026-10`, rolled 2026-10-01; September closed 1.50 / 3.50, history kept verbatim — the `spent:` note carries the open `.de` renewal question, REPORT.md §14) with `"warnings": []` while the month is current, and `GLADEX_CURRENT_MONTH=2099-12` → still exit **0** with exactly one warning; `/api/budget` on :8000/:8001 → same numbers, both pages `200` (PHP parses the ledger independently)
264:- Live (measured 2026-10-03): `dns-verify gladex.de` → exit 0, `gladex.de BOGUSTYPE` → exit 2 (was 0 + fabricated record), `--server 203.0.113.1 --timeout 1` → exit 4 (was 3), `this-domain-should-be-available-12345.de` → exit 3 (no records), `--format human gladex.de NS` → human-readable NS records (`Domain: gladex.de` / `Queried: NS` / `Has records: Yes`)
841:  Live transcript (real `repo-lint` child, real tool, throwaway repo): the changelog
929:- Live (2026-09-25, after this step): `promote-dev-to-prod --dry-run --force --format json` →
936:- Live (measured 2026-10-03): `promote-dev-to-prod --dry-run` → exit 5, `GATE REFUSED (exit 5): verdict VERDICT-20260930T022500-team-sofia.md is STALE (promoted source [app/src/php, examples/workflows] changed …s after it — re-review required)` — the *third* of the verdict gate's three exit-5 refusals (an empty `mailboxes/reviewer-to-main/` is the first, a newest verdict that is not APPROVE the second); this bullet claimed the empty-queue reason until 2026-10-03, when that directory held five verdict files and the refusal came from staleness instead. The refused run reports `prod untouched`.
937:- Live (measured 2026-10-10): `promote-dev-to-prod --ledger-seed` → exit 0, `seeded
954:- Live (2026-09-25, `[0.4.47]`): `promote-dev-to-prod --dry-run --force --format json`
1045:- Live (measured 2026-10-03): `ip-drift-check --format json --domain gladex.de` → exit 0, `drift: false` with `public_ip 77.90.15.49`, `dns_records ["77.90.15.49"]`, `dns_status: "ok"` (tunnel up); `--help` → 0.
1102:- **Live (real zone, 2026-09-24)**: `zones` → `200 ["gladex.de."]`; `records` → the real rrset listing; TXT probe `set-txt gladex.de _agentprobe …` → `204 OK`, stored as `"…"`, **`dig +short TXT _agentprobe.gladex.de` resolved it**, then `delete … TXT` → `204 OK` and gone. This run is also what exposed that the live API answers `/zones` with a **bare array**, not the documented `{"zones": [...]}` object — the shape the fake server had wrongly modelled, now covered by `zones`/`wrapped`/`scalar` scenarios. Live gate proofs: `delete … NS`, `set-a other.de …`, `set-a … 86400` all exit 2 without contacting the API.
> build · mimo-v2.6-flash-free
$ cd /data/repo && sed -n '8235,8340p' tools/REGISTRY.md
## shift-tally

**Purpose**: Read-only summariser over the persona transcripts' shift footers
(`agent-logs/identity-<name>.md`, one append-only log per persona runner): a
per-model and a per-identity table of total / `exit=0` / `exit=1` / failure
rate, using the marker shape the live `failed-units` row reads. It is the
durable form of run 917's throwaway measurement (430 counted shifts over 19
days) and the reader behind the standing census that both red identity units
ran `ling-3.1-flash-free`: it turns that one-off number into something any run
can re-read. No write mode, no truncate, no unlink, no network, no child
process — the transcripts are the persona runners' own logs and this tool never
writes one.

**Location**: `/data/repo/tools/shift-tally`

**Usage**:
```
shift-tally [--format {json,human}] [--dedup {file,global,both}]
```

**Options**:
- `--format {json,human}` - Output format (default: human)
- `--dedup {file,global,both}` - Which de-duplication level(s) to report (default: both)
- `--help` - Show this help message

**Marker + gates**: `===== shift <date -Is> model=<model> exit=<n> =====`,
**anchored** (nothing else on the line), the timestamp parsable, and not more
than **300 s** ahead of the clock — the `[0.4.307]` hardened shape and gates
1–3 verbatim, so the tool counts the same population the live read counts. A
line containing `===== shift ` that does not match the shape is counted in
`not_anchored` (the exclusion class `[0.4.307]` already makes, surfaced rather
than silent) and skipped; an unparsable timestamp is `bad_ts`, a future-dated
one `future_ts`.

**De-duplication**: `file` keeps the last marker per timestamp inside each
transcript; `global` additionally de-duplicates on `(ts, model, code)` across
all transcripts (the conservative read, which absorbs a persona quoting another
identity's marker); `both` prints both. The two agree to within ~2 points on
every model — the reason both are offered rather than one being chosen silently.

**Exit codes**:
- `0` - ok: the transcript directory was read and summarised
- `1` - unexpected internal error
- `2` - invalid arguments: unknown flag, missing value, `--format` outside json|human, or `--dedup` outside file|global|both — argparse validates before any I/O
- `3` - cannot verify: the transcript directory is missing or holds no persona transcript — never a pass; the refusal rides in the exit code. Scope of 3: it is per directory, the single unit this tool reads.

**JSON keys** (one object per requested de-duplication level): `tool`, `ts`, `dir`, `dedup`, `ok`, `files`, `shift_lines`, `anchored`, `not_anchored`, `bad_ts`, `future_ts`, `total`, `exit0`, `exit1`, `rate`, `first_ts`, `last_ts`, `models`, `identities`, `errors`.

**File selection**: a persona transcript is `identity-<name>.md` with a single
lowercase word for the name; a dated/hyphenated sibling
(`identity-shift-red-modes-...md`) is a report, not a transcript, and is
excluded by that line rather than a hand list (measured 2026-10-12: it holds
**0** anchored markers, so the rule and the content agree).

**Test hook**: `GLADEX_SHIFT_TALLY_DIR` — reads `<DIR>/identity-*.md` instead
of `/data/repo/agent-logs`. It is an environment variable rather than a flag
pair on purpose: the advertised option surface stays the two pairs above, which
is the surface `tests/test_registry_coverage.sh` probes. Unset in normal use.

**Verification (2026-10-12, run 918)**: run against the live seven transcripts
it reproduces run 917's `agent-logs/model-tally-2026-10-12.md` exactly — the
`file` read is that document's Table A (**452 shifts / 363 ok / 89 fail /
19.69 %**, `nemotron-3-ultra` 190/176/14 = 7.37 %, `ling-3.1` 76/54/22 =
28.95 %, `mimo` 50/26/24 = 48.00 %) and the `global` read its Table B (**430 /
346 / 84 / 19.53 %**, `nemotron-3-ultra` 186/172/14 = 7.53 %, `ling-3.1`
74/54/20 = 27.03 %), window `2026-09-23T20:11:29+02:00 … 2026-10-12T14:17:48+02:00`,
corpus **1,488** shift-token lines / **1,060** anchored / **428** not-anchored /
**0** bad-ts / **0** future-ts. A dedicated `tests/` suite is a deliberate
follow-up (it would be the 104th and would move the `- Live:` figure); this
run's step is the reader, the registry section and the C28g scope refresh.
*(That follow-up landed the same day: `tests/test_shift_tally.sh`, the 104th
suite, moved the `- Live:` figure **103 → 104** and is documented in
`[0.4.319]`.)*

**Dependencies**: python3 (stdlib only: argparse, datetime, glob, json, os, re,
sys).

**Status**: Built ✅ 2026-10-12 (a sibling desk's work, swept into `main` by
`8fb4270` at 16:51 CEST and documented in `[0.4.318]`, which takes run 917's
queued durable-tally follow-up) — registers with that suite's C28g option-list scope
**26 → 27**; its own `tests/` guard (`tests/test_shift_tally.sh`, the 104th
suite, `[0.4.319]`) moved the `- Live:` figure **103 → 104**. G38 `cl_other`
stays **25**: neither new pair lands in it — `--format` takes the fmt bucket and
`--dedup {file,global,both}` the enum bucket — and the usage row names both
pairs verbatim, which is what (79)/G33 reads.
> build · mimo-v2.6-flash-free
$ cd /data/repo && grep -n 'A28\|A12\|A30\|A3\b\|A15' tools/red-watch | head -40
23:    reader can tell "another identity's untracked file (A3)" from "repo-lint
> build · mimo-v2.6-flash-free
$ cd /data/repo && grep -rn 'A28' --include='*.sh' --include='*.md' --include='red-watch' tools/ tests/ agent-logs/ 2>/dev/null | grep -v 'agent-logs/identity-' | head -20
tools/REGISTRY.md:6147:  them read as *not* waited (**A28**, so arity cannot over-correct into "an
tools/REGISTRY.md:6289:  (`A15 A27 A28 A29 A38 A39 A40 A41 A42 A43 A44 B14 B24 B27 B30`)**, 140/15 —
tools/REGISTRY.md:6307:  (`A15 A27 A28 A29 A38 A39 A40 A41 B14 B24 B27`)**, 64/11 — A41 and B27 are
tools/REGISTRY.md:6318:  (`A15 A27 A28 A29 A38 A39 A40 B14 B24`)**, 61/9 — it asks only whether a
tools/REGISTRY.md:6366:  (`A15 A27 A28 A29 B14`)**, 45/5 — one row *changed*, because every (86)
tools/REGISTRY.md:7417:`docs`, `A28`/`A29`), i.e. both detectors item (114) named, through one
tests/test_detached_children.sh:627:# could pass a wrong one); A28 keeps the arity fix from over-correcting into
tests/test_detached_children.sh:643:assert_false "A28 plant: three children and a wait naming two of them read as NOT waited — arity is bounded by the names, not by 'a wait covers everything'" $?
tests/test_detached_children.sh:1221:  ctl_run K6 "A15 A27 A28 A29 A38 A39 A40 A41 A42 A43 A44 B14 B24 B27 B30" "$d"
tests/test_regression_run.sh:393:# A26-A28: --version flag (added 2026-10-10)
tests/test_regression_run.sh:398:assert_eq "A28 --version --format json is valid JSON with tool+version" \
tests/test_monitor_cascade.sh:158:# `base..HEAD` is empty (A28 can hold). Both are what make section A's
tests/test_gladex_monitor.sh:249:add A28 rc          docs       0                                            "no undocumented code commit since the last CHANGELOG commit"
agent-logs/PROGRESS.md:7225:- **All thirteen mutation controls re-run in sandbox copies at 50 assertions** (`/tmp/opencode/controls86.py`, full copies of `tests/` + `tools/`, every patch asserted to have applied before the suite ran, live tree never written, untouched copy **50/0**). **Four new**: counting reader back → **3 red (`A26 A27 B14`)**, 47/3; identity resolution removed → **2 red (`A27 B14`)**, 48/2; arity removed (one cover per `wait`) → **1 red (`A26`)**, 49/1; operator cut removed → **1 red (`A29`)**, 49/1. **Nine older rows**: `setsid` plant **3 (`A2 C1 C2`)**, `&` plant **1 (`A4`)**, real `wait` commented **1 (`A5`)**, guard tokens renamed **2 (`A3 C1`)**, over-strict bare `wait` **1 (`A16`)**, `cmd-str` removed **2 (`A18 A23`)**, naive `source` **2 (`A19 A25`)** — seven unchanged — and **pre-(84) last-vs-last 1 red → 5 red (`A15 A27 A28 A29 B14`)**, 45/5, which is the one historical figure that *moved*: every (86) plant and B14 puts a `wait` after the file's last `&`, the only question that reader asks. The 42-assertion figure stays as written in `[0.4.144]`; 5 is the measured number now, and `tools/REGISTRY.md` says both.
agent-logs/PROGRESS.md:7455:- **Measured, not claimed — 50 → 102 in 10.4 s → 2 m 27 s**: `bash tests/test_detached_children.sh` → **102 passed, 0 failed**, counted by tallying `PASS:` lines per section (**A 30, B 17, C 3, K 52**), against **50 / 0 (10.4 s)** at `cee7334` before the step — **52 added, 0 removed**. The thirteen measured red sets, each asserted exactly and in order: `A2 C1 C2` · `A4` · `A5` · `A3 C1` · *(none)* · `A15 A27 A28 A29 B14` · `A16` · `A18 A23` · `A19 A25` · `A26 A27 B14` · `A27 B14` · `A26` · `A29` — **no figure in `REGISTRY.md`'s control table had to change**, i.e. the prose that had never been executed was right on all thirteen rows, and now nothing can let it stop being right.
agent-logs/PROGRESS.md:7488:- **Three corrections the first full run reddened on — each red named a helper or a reader, never an A/B/C assertion, and nothing was loosened to reach green.** Run 1 → **111 passed / 5 failed (3 m 8 s)**: (i) `K14a/c/d` — **`awk -v pat=` interprets backslash escapes**, so K14's needles (`wait([ \t;|]|$)`) travelled as a real TAB: `has_pre` answered `no` on the *live* file (the widened class it was looking for → `got='no no'`) and `ctl_replace` exited 1, so the copy stayed unmutated and ran green on nothing; `has_pre`, `ctl_replace` and `ctl_delete` now carry `pat`/`rep` in **`ENVIRON`**, where no escape is read (`$K_MARK` stays with `-v`, it has no backslash). (ii) `K12c` → `A15 A27 A28 A29 B14` — the tail handed back a **whitespace** terminator too, so `split` produced a leading EMPTY token and K12's capped loop (`i <= 1`) counted nothing → `nargs == 0` → `cover_all()` → every argument-bearing wait read bare; only `;`/`|` are handed back now (whitespace/EOL behave exactly as before) → `K12c` = `A26`. (iii) `K10c` → `A26 A27 A32 B14` against a want of `A26 A27 B14` — measured, so the **want was corrected, not the code**: the pre-(86) counting reader never had (86)'s operator cut either, so `wait; echo done` reads `; echo done` as three arguments, `nargs ≠ 0`, one decrement over three children, two left pending — the fourth red is the counting reader's own and is written down in the control's comment (the cut is K13's subject on the live reader).
agent-logs/PROGRESS.md:7490:- **Measured, not claimed — 102 → 116 in 3 m 6 s**: `bash tests/test_detached_children.sh` → **116 passed, 0 failed**, counted by tallying `PASS:` lines per section (**A 33, B 20, C 3, K 60**), against **102 / 0 (2 m 27 s)** at `[0.4.151]` — **14 added, 0 removed** (A 30→33, B 17→20, K 52→60, C 3 untouched), the child suite pinned at **56** by every `Kn b` (was 50). The fifteen measured red sets, each asserted exactly and in order: `A2 C1 C2` · `A4` · `A5` · `A3 C1` · *(none)* · `A15 A27 A28 A29 B14` · `A16 A32` · `A18 A23` · `A19 A25` · `A26 A27 A32 B14` · `A27 B14` · `A26` · `A29` · `A30 A31 A32 B18` · `A32`. Two of the thirteen (100)-era rows moved with the new plants, both predicted before they were run: **K7** `A16` → **`A16 A32`** (a bare wait closed by a separator over three children is the over-correction's second instance) and **K10** `A26 A27 B14` → **`+ A32`** (above).
agent-logs/PROGRESS.md:7602:- **The baseline, measured on the shipped tree before a line was written.** On `09575e9` (the commit that carried run 436's widening, i.e. `a726d49`'s content in `main`), `bash tests/test_detached_children.sh` → **133 passed / 5 failed**, the five reds exactly and only `K6c K7c K10c K13c K15c` — no `A`, no `B`, no `C` red, so the widened reader itself was already correct and only the controls' documented sets were stale. Each red's own line is the measurement (`/tmp/opencode/m104b/run-base.log`): `K6c want='A15 A27 A28 A29 B14' got='A15 A27 A28 A29 A38 A39 A40 B14 B24'`; `K7c want='A16 A32' got='A16 A32 A37'`; `K10c want='A26 A27 A32 B14' got='A26 A27 A32 A37 A38 A39 A40 B14 B24'`; `K13c want='A29' got='A29 A37 A40'`; `K15c want='A32 A37' got='A32'`. The first four are sets that grew by exactly the plants (104) added — the subshell rule's three plants `A38 A39 A40` plus `B24`, and `A37`'s `&&`-closed bare wait — so the reader grew and the expectation did not. The fifth went the other way, and that direction is the finding.
agent-logs/PROGRESS.md:7604:- **What landed: five `ctl_run` want-strings, their derivation comments, one corrected reader comment.** In `tests/test_detached_children.sh` the five sets are rewritten to the measured ones — `K6` → `A15 A27 A28 A29 A38 A39 A40 B14 B24`, `K7` → `A16 A32 A37`, `K10` → `A26 A27 A32 A37 A38 A39 A40 B14 B24`, `K13` → `A29 A37 A40`, `K15` → `A32` — and each control's comment gains the reason in the control's own voice rather than a bare number: `K6` asks **one** question (is a WAIT record after the file's last `&`?) which a pipeline's wait satisfies, so it has no subshell rule to lose and nothing else in it *can* redden; `K7`'s three reds are three instances of one subject (a bare wait over three children, closed by `;`, by `&&`, spaced); `K10`'s five additions are each a rule that reader never had (`A37` leaves `&&echo` to the `&` filter and `done` as the one argument; `A38 A39 A40`/`B24` have no subshell rule); `K13`'s `A40` is the cut's own casualty, because `op == "|"` is what flags a left-hand subshell wait and `op` is set **by** the cut, with `head` unable to see the `|` behind the match. The reader comment above the `BGWAIT_AWK` token loop is corrected where it had been **false**: it credited `&` joining the class with `A37`, and `K15` proved that credit wrong — with the tail eating the separator the list is emptied anyway, by the operator cut firing on the leading `&`, so `A37` does not join and the `&` half of the hand-back class is recorded **UNPROVEN** rather than covered.
agent-logs/PROGRESS.md:7708:- **What landed (`tests/test_detached_children.sh` only).** The reader in both places (pairing + census): `off = (mt ~ /[;|]$/) ? RLENGTH - 1 : RLENGTH` — `&` no longer handed back — with the surrounding comments rewritten from *"unproven, queued (107)"* to what was measured. Fixture **`wait_amp_cmd`**, created **before** the census so the plant is counted rather than discovered, plus **A41** (reads as *not* waited); three-child fixture **`fix/f7.sh`** (`f7a`/`f7b`/`f7c`) plus the process half **B27** (verdict on that very file: *not* waited), **B28** (exits **0 inside the `timeout 5`** bound — nothing blocked, `wait&` starts the wait as a job), **B29** (child `F7A` **still alive**); those three pids joined `B10`'s reap census; **C4** counts backticks on every `assert_*`/`ok`/`fail` line and reds at anything but 0; `ctl_run`'s child pin **70 → 75**; **K6**'s documented set **→ 11 red** (`A15 A27 A28 A29 A38 A39 A40 A41 B14 B24 B27`, 64/11); header *"thirteen … / fifteen"* clause repaired to **seventeen**.
Error: Rate limit exceeded. Please try again later.
exit=1

Select another run log from the list above. Only files matching run-YYYYMMDD-HHMMSS-N.log are readable.

[Mon Oct 12 20:48:37 2026] 127.0.0.1:56948 Accepted
[Mon Oct 12 20:48:37 2026] 127.0.0.1:56948 Closing
[Mon Oct 12 20:55:34 2026] 127.0.0.1:36618 Accepted
[Mon Oct 12 20:55:34 2026] 127.0.0.1:36618 Closing
[Mon Oct 12 20:55:35 2026] 127.0.0.1:54530 Accepted
[Mon Oct 12 20:55:35 2026] 127.0.0.1:54530 Closing
[Mon Oct 12 20:55:36 2026] 127.0.0.1:54542 Accepted
[Mon Oct 12 20:55:36 2026] 127.0.0.1:54542 Closing
[Mon Oct 12 20:55:38 2026] 127.0.0.1:54550 Accepted
[Mon Oct 12 20:55:38 2026] 127.0.0.1:54550 Closing
[Mon Oct 12 20:55:40 2026] 127.0.0.1:54554 Accepted
[Mon Oct 12 20:55:40 2026] 127.0.0.1:54554 Closing
[Mon Oct 12 20:55:45 2026] 127.0.0.1:45430 Accepted
[Mon Oct 12 20:55:45 2026] 127.0.0.1:45430 Closing
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58352 Accepted
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58352 Closing
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58366 Accepted
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58366 Closing
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58372 Accepted
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58372 Closing
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58386 Accepted
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58386 Closing
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58388 Accepted
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58388 Closing
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58394 Accepted
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58394 Closing
[Mon Oct 12 21:00:19 2026] 127.0.0.1:58404 Accepted
[Mon Oct 12 21:00:20 2026] 127.0.0.1:58404 Closing
[Mon Oct 12 21:00:20 2026] 127.0.0.1:58420 Accepted
[Mon Oct 12 21:00:20 2026] 127.0.0.1:58420 Closing
[Mon Oct 12 21:00:20 2026] 127.0.0.1:58428 Accepted
[Mon Oct 12 21:00:20 2026] 127.0.0.1:58428 Closing
[Mon Oct 12 21:00:20 2026] 127.0.0.1:58432 Accepted
[Mon Oct 12 21:00:20 2026] 127.0.0.1:58432 Closing
[Mon Oct 12 21:01:14 2026] 127.0.0.1:41880 Accepted
[Mon Oct 12 21:01:14 2026] 127.0.0.1:41880 Closing
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43776 Accepted
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43776 Closing
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43782 Accepted
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43782 Closing
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43796 Accepted
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43796 Closing
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43812 Accepted
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43812 Closing
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43826 Accepted
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43826 Closing
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43830 Accepted
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43830 Closing
[Mon Oct 12 21:01:25 2026] 127.0.0.1:43844 Accepted
[Mon Oct 12 21:01:26 2026] 127.0.0.1:43844 Closing
[Mon Oct 12 21:01:26 2026] 127.0.0.1:43854 Accepted
[Mon Oct 12 21:01:26 2026] 127.0.0.1:43854 Closing
[Mon Oct 12 21:01:26 2026] 127.0.0.1:43870 Accepted
[Mon Oct 12 21:01:26 2026] 127.0.0.1:43870 Closing
[Mon Oct 12 21:01:26 2026] 127.0.0.1:43882 Accepted
[Mon Oct 12 21:01:26 2026] 127.0.0.1:43882 Closing
[Mon Oct 12 21:05:05 2026] 127.0.0.1:60140 Accepted
[Mon Oct 12 21:05:05 2026] 127.0.0.1:60140 Closing
[Mon Oct 12 21:05:05 2026] 127.0.0.1:60156 Accepted

Other app logs: downloads.log