diff --git a/agents/researcher/work/queue-71/r1/BUILD.md b/agents/researcher/work/queue-71/r1/BUILD.md
new file mode 100644
index 000000000..ebfacf5b4
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/BUILD.md
@@ -0,0 +1,83 @@
+# Row 71 (#1565) build, round 1
+
+Phase B done. Gate passed; both mutants killed. Nothing committed, queue row not moved, no Gitea or tracker writes.
+
+## Identity
+
+- Row 71: tree read follow-up, confirm-read tests and the fake's view 404.
+- Brief: `docs/plans/2026-10-11_tree-read-follow-up.md` (not edited; blob 499c6cdc).
+- Owner: researcher (queue owner; decisions 122 and 129). Reviewers: Dewey and Filbert. Sage runs the landing gate.
+- Base: `ac6b56ee55278077d61f39e6ca42513725427154` (`base.txt`).
+- Worktree: `.claude/worktrees/agent-a44468a48c485520e`, detached at the base, candidate uncommitted.
+
+## Changes
+
+`packages/tasks/src/fake.mjs`
+- A missing kanban view answers 404 with code 3014, not 13001, in both `_buckets` and `_board`. The detail string is unchanged.
+- The `deleteView` comment cites the live record `agents/sage/work/vikunja-setup/2026-10-11_tick-62.json` (`deletedViewProbe`).
+
+`packages/tasks/tests/tree.test.mjs`
+- Test 18 comment says 3014 and cites the same record.
+- Three steps added after the existing "404 on the list, 503 on the project read" step. Each sets the child's task list to 404, then the confirm read to one of three answers. Each expects `tracker-not-found` and a refused tick with the child kept:
+  - confirm read 429 (Filbert N3, `r3-confirm-4xx`)
+  - confirm read 401 (Filbert N3)
+  - confirm read status 0, a failed call (Dewey, `r3-confirm-error-on-throw`)
+
+`build.patch` covers these two paths only (64 lines, `git diff --binary` against the base). It applies cleanly to the base with `git apply --cached --check` on a temporary index.
+
+## Rulings applied
+
+1. Dewey's mutant is the status-0 form. Both mutants were defined from the brief; neither text exists in the repo. They are in `mutants.py`.
+2. Owner: built as researcher. The brief names Darkwing as owner. The brief is not edited, and the mismatch is recorded here.
+3. The fake keeps the existing detail string.
+4. A view under another project: the fake returns 3014 for both cases. That case was never read live, and the fake does not separate it.
+
+## Deviations and notes
+
+- Rebase: the two edits were set aside as a tagged stash entry, checked out at the base, reapplied by SHA, and the entry was dropped by its ref. Other stash entries were not touched.
+- Node suites run with `--test-reporter=tap` so the counts parse. They are otherwise the commands Sage named.
+- Only `test-task.sh` gets `DOCKER_HOST=unix:///nonexistent.sock`. The node suites get it too, as ruled.
+- Test 18 stops at its first failing assertion. The 401 step is not isolated by a named mutant. The 4xx mutant would also fail it, but the run stopped at the 429 step.
+- The deleted-view case itself was not read live. The probe in tick-62 used a view id that no view has, as the brief notes.
+- No baseline run against the pre-change tests. The claim that the two mutants survive the old tests is Filbert's and Dewey's, not reproduced here.
+
+## Gate (steps 1 and 2)
+
+Run by `gate.sh` from the worktree root, chained by `chain.sh`. Each child ran in its own session, under `run_group.py`, which kills the process group on timeout. No timeouts occurred. Manifest check at gate start: both files OK.
+
+| step | command | result | UTC start-end |
+| --- | --- | --- | --- |
+| node tasks | `node --test packages/tasks/tests/*.test.mjs` | 89 tests, 89 pass, 0 fail, rc 0 | 14:46:36-14:46:37 |
+| node bus | `node --test packages/bus/tests/*.test.mjs` | 115 tests, 115 pass, 0 fail, rc 0 | 14:46:37-14:46:38 |
+| test-auth | `scripts/test-auth.sh` | selftest 15 passed, 0 failed | 14:46:38 |
+| test-conductor | `scripts/test-conductor.sh` | selftest 17 passed, 0 failed | 14:46:38-14:46:44 |
+| test-config | `scripts/test-config.sh` | selftest 24 passed, 0 failed | 14:46:44-14:46:46 |
+| test-discord | `scripts/test-discord.sh` | discord suite 66 passed, 0 failed | 14:46:46-14:46:53 |
+| test-extension-package | `scripts/test-extension-package.sh` | selftest 18 passed, 0 failed | 14:46:53-14:46:54 |
+| test-foundation | `scripts/test-foundation.sh` | selftest 44 passed, 0 failed | 14:46:54-14:47:32 |
+| test-queue | `scripts/test-queue.sh` | queue suite 27 passed, 0 failed | 14:47:32-14:47:52 |
+| test-release | `scripts/test-release.sh` | selftest 14 passed, 0 failed | 14:47:52-14:48:02 |
+| test-task | `scripts/test-task.sh` (DOCKER_HOST dead) | selftest 26 passed, 0 failed | 14:48:02-14:48:04 |
+
+Gate PASS at 14:48:04 UTC. The script suites print their own pass lines, not TAP, so the summary holds only the exit status for them. The full logs are in `out/`.
+
+## Mutants (steps 3, after gate PASS)
+
+Each mutant was applied alone to `packages/tasks/src/sync.mjs`, then the tasks suite ran. The file was restored from bytes held in memory and checked after each restore.
+
+| mutant | edit to sync.mjs | result | killing test | failing assertion |
+| --- | --- | --- | --- | --- |
+| `r3-confirm-4xx` | `return r.status >= 400 && r.status < 500;` | killed (rc 1; 89 run, 88 pass, 1 fail) | test 18, "a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only" | `tree.test.mjs:465`, the 429 confirm step |
+| `r3-confirm-error-on-throw` | `return r.status === 403 \|\| r.status === 404 \|\| r.status === 0;` | killed (rc 1; 89 run, 88 pass, 1 fail) | same test | `tree.test.mjs:472`, the status-0 confirm step |
+
+Mutant window: 14:48:04-14:48:06 UTC. After restore, `sync.mjs` sha256 is `35e79dc2ac56af8047bef11a5aaf5f1d37ce3e11513f9968454053068f59ebcf`, the original, and `git diff` shows no change to it.
+
+## Files in this packet
+
+- `base.txt`, `build.patch`, `files.txt`, `candidate-manifest.sha256`
+- `run_group.py` (process-group runner), `gate.sh` (steps 1 and 2), `chain.sh` (gate, then mutants on pass), `mutants.py` (steps 3)
+- `out/`: `gate-summary.txt`, `gate-stdout.txt`, `manifest-check.txt`, `node-tasks.log`, `node-bus.log`, `test-*.log`, `mutant-*.log`, `mutants-summary.txt`, `mutants-stdout.txt`, `chain-start.txt`, `chain-rc.txt`
+
+## Process check
+
+No test, node or chromium process of mine remains. Checked at 14:48 UTC by scanning `/proc` cwd links inside this worktree. The one leftover was my own `tail -F` monitor, stopped.
diff --git a/agents/researcher/work/queue-71/r1/base.txt b/agents/researcher/work/queue-71/r1/base.txt
new file mode 100644
index 000000000..8d25787e1
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/base.txt
@@ -0,0 +1 @@
+ac6b56ee55278077d61f39e6ca42513725427154
diff --git a/agents/researcher/work/queue-71/r1/build.patch b/agents/researcher/work/queue-71/r1/build.patch
new file mode 100644
index 000000000..8652afe5e
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/build.patch
@@ -0,0 +1,64 @@
+diff --git a/packages/tasks/src/fake.mjs b/packages/tasks/src/fake.mjs
+index 633224db8..57314be46 100644
+--- a/packages/tasks/src/fake.mjs
++++ b/packages/tasks/src/fake.mjs
+@@ -138,7 +138,8 @@ export class FakeVikunja {
+     for (const id of gone) this.#projects.delete(id);
+   }
+   // Drops a view, its buckets and the tasks' places in them. The project still reads, and a read
+-  // of the view answers 404 with code 13001.
++  // of the view answers 404 with code 3014, as tasks.woltje.com v2.7.0 does (Sage's live probe,
++  // agents/sage/work/vikunja-setup/2026-10-11_tick-62.json, deletedViewProbe).
+   deleteView(view) {
+     for (const [id, b] of this.#buckets) if (b.view === view) this.#buckets.delete(id);
+     for (const t of this.#tasks.values()) t.buckets.delete(view);
+@@ -473,14 +474,14 @@ export class FakeVikunja {
+   _buckets(user, [p, k], q) {
+     const r = this.#projectFor(user, p);
+     if (r.error) return r.error;
+-    if (this.#views.get(k)?.project !== p) return { status: 404, body: { title: 'Not Found', status: 404, detail: 'This project view does not exist.', code: 13001 } };
++    if (this.#views.get(k)?.project !== p) return { status: 404, body: { title: 'Not Found', status: 404, detail: 'This project view does not exist.', code: 3014 } };
+     return { status: 200, body: page([...this.#buckets.values()].filter((b) => b.view === k).map((b) => this.#bucketJson(b)), q) };
+   }
+   _board(user, [p, k], q) {
+     const r = this.#projectFor(user, p);
+     if (r.error) return r.error;
+     const v = this.#views.get(k);
+-    if (v?.project !== p) return { status: 404, body: { title: 'Not Found', status: 404, detail: 'This project view does not exist.', code: 13001 } };
++    if (v?.project !== p) return { status: 404, body: { title: 'Not Found', status: 404, detail: 'This project view does not exist.', code: 3014 } };
+     const f = q.get('filter') ?? '';
+     if (f && f !== 'done = false') return { status: 400, body: bad('unsupported filter in the fake') };
+     const per = Math.min(Number(q.get('per_page')) || 50, 50),
+diff --git a/packages/tasks/tests/tree.test.mjs b/packages/tasks/tests/tree.test.mjs
+index 5ac99ae51..e325be398 100644
+--- a/packages/tasks/tests/tree.test.mjs
++++ b/packages/tasks/tests/tree.test.mjs
+@@ -430,8 +430,8 @@ test('a child whose kanban view is deleted still refuses the tick, and the recon
+   const rootTask = await add(w, w.project, 'root');
+   await ready(w);
+   const rows = () => w.store.get('SELECT (SELECT count(*) FROM events) + (SELECT count(*) FROM task_snapshots) n').n;
+-  // The board answers 404 with code 13001, and the project still reads, so the child hasn't left
+-  // the tree. The tick reads the project, fails, and writes nothing, not even the root's edit.
++  // The board answers 404 with code 3014 (the live v2.7.0 answer, tick-62 record), and the project
++  // still reads, so the child hasn't left the tree. The tick reads the project, fails, and writes nothing, not even the root's edit.
+   w.fake.deleteView(w.fake.kanban(child).id);
+   await w.ui('PATCH', `/tasks/${rootTask}`, { title: 'root renamed' });
+   await sleep(20);
+@@ -459,6 +459,17 @@ test('a child whose kanban view is deleted still refuses the tick, and the recon
+   w.fake.fault('GET', new RegExp(`^/projects/${child}/tasks$`), 404);
+   w.fake.fault('GET', new RegExp(`^/projects/${child}$`), 503);
+   await assert.rejects(w.adapter.tick('demo'), { code: 'tracker-not-found' });
++  // The confirm read answers another 4xx: the tick still refuses, and the child stays.
++  w.fake.fault('GET', new RegExp(`^/projects/${child}/tasks$`), 404);
++  w.fake.fault('GET', new RegExp(`^/projects/${child}$`), 429);
++  await assert.rejects(w.adapter.tick('demo'), { code: 'tracker-not-found' });
++  w.fake.fault('GET', new RegExp(`^/projects/${child}/tasks$`), 404);
++  w.fake.fault('GET', new RegExp(`^/projects/${child}$`), 401);
++  await assert.rejects(w.adapter.tick('demo'), { code: 'tracker-not-found' });
++  // The confirm read's call fails (status 0, which the client returns for a thrown fetch): the tick refuses.
++  w.fake.fault('GET', new RegExp(`^/projects/${child}/tasks$`), 404);
++  w.fake.fault('GET', new RegExp(`^/projects/${child}$`), 0);
++  await assert.rejects(w.adapter.tick('demo'), { code: 'tracker-not-found' });
+   await w.ui('PATCH', `/tasks/${task}`, { title: 'child renamed' });
+   await sleep(20);
+   await w.adapter.tick('demo');
diff --git a/agents/researcher/work/queue-71/r1/candidate-manifest.sha256 b/agents/researcher/work/queue-71/r1/candidate-manifest.sha256
new file mode 100644
index 000000000..283f549e5
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/candidate-manifest.sha256
@@ -0,0 +1,2 @@
+feb4a6f8927471ce445082d5a28e0e29377b000aabec97ea14378dafe88ec065  packages/tasks/src/fake.mjs
+bc081ac487671bfb79c9afbc5bb9788560d1643f1975fa92bb31865022b2cba9  packages/tasks/tests/tree.test.mjs
diff --git a/agents/researcher/work/queue-71/r1/chain.sh b/agents/researcher/work/queue-71/r1/chain.sh
new file mode 100644
index 000000000..e505e5298
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/chain.sh
@@ -0,0 +1,12 @@
+#!/bin/bash
+# Row 71 round 1 chain: the gate (steps 1 and 2), then the mutants only if the gate exits 0.
+# Run from the worktree root.
+P=agents/researcher/work/queue-71/r1
+mkdir -p "$P/out"
+date -u > "$P/out/chain-start.txt"
+bash "$P/gate.sh" > "$P/out/gate-stdout.txt" 2>&1 \
+  && python3 "$P/mutants.py" > "$P/out/mutants-stdout.txt" 2>&1
+rc=$?
+echo "chain rc=$rc" > "$P/out/chain-rc.txt"
+date -u >> "$P/out/chain-rc.txt"
+exit $rc
diff --git a/agents/researcher/work/queue-71/r1/files.txt b/agents/researcher/work/queue-71/r1/files.txt
new file mode 100644
index 000000000..dbc10ff26
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/files.txt
@@ -0,0 +1,2 @@
+packages/tasks/src/fake.mjs
+packages/tasks/tests/tree.test.mjs
diff --git a/agents/researcher/work/queue-71/r1/gate.sh b/agents/researcher/work/queue-71/r1/gate.sh
new file mode 100644
index 000000000..2a025f705
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/gate.sh
@@ -0,0 +1,45 @@
+#!/bin/bash
+# Row 71 round 1 gate, run from the worktree root (base ac6b56ee5 plus build.patch).
+# Steps chain with && and stop at the first failure. Each child runs through run_group.py,
+# which kills its whole process group on timeout.
+P=agents/researcher/work/queue-71/r1
+OUT=$P/out
+RUN="python3 $P/run_group.py"
+S=$OUT/gate-summary.txt
+mkdir -p "$OUT"
+: > "$S"
+stamp() { date -u +%Y-%m-%dT%H:%M:%SZ; }
+note() { echo "$*" >> "$S"; }
+
+# step NAME TIMEOUT_S LOGNAME -- CMD...
+step() {
+  local name=$1 t=$2 log=$OUT/$3 rc t0
+  shift 3
+  shift # the --
+  t0=$(stamp)
+  $RUN "$t" "$log" "$PWD" -- "$@"
+  rc=$?
+  note "$name rc=$rc start=$t0 end=$(stamp) $(grep -E '^# (tests|pass|fail|cancelled|skipped|todo) ' "$log" | tr -s ' ' | tr '\n' ' ')"
+  [ "$rc" -eq 0 ] || note "$name FAILED; last line: $(grep -v '^\s*$' "$log" | tail -1 | cut -c1-200)"
+  return "$rc"
+}
+
+note "gate start $(stamp) base $(git rev-parse HEAD)"
+if sha256sum -c "$P/candidate-manifest.sha256" > "$OUT/manifest-check.txt" 2>&1 \
+  && step node-tasks 900 node-tasks.log -- env DOCKER_HOST=unix:///nonexistent.sock node --test --test-reporter=tap 'packages/tasks/tests/*.test.mjs' \
+  && step node-bus 900 node-bus.log -- env DOCKER_HOST=unix:///nonexistent.sock node --test --test-reporter=tap 'packages/bus/tests/*.test.mjs' \
+  && step test-auth 1200 test-auth.log -- env -u DOCKER_HOST bash scripts/test-auth.sh \
+  && step test-conductor 1200 test-conductor.log -- env -u DOCKER_HOST bash scripts/test-conductor.sh \
+  && step test-config 1200 test-config.log -- env -u DOCKER_HOST bash scripts/test-config.sh \
+  && step test-discord 1200 test-discord.log -- env -u DOCKER_HOST bash scripts/test-discord.sh \
+  && step test-extension-package 1200 test-extension-package.log -- env -u DOCKER_HOST bash scripts/test-extension-package.sh \
+  && step test-foundation 1200 test-foundation.log -- env -u DOCKER_HOST bash scripts/test-foundation.sh \
+  && step test-queue 1200 test-queue.log -- env -u DOCKER_HOST bash scripts/test-queue.sh \
+  && step test-release 1200 test-release.log -- env -u DOCKER_HOST bash scripts/test-release.sh \
+  && step test-task 1200 test-task.log -- env DOCKER_HOST=unix:///nonexistent.sock bash scripts/test-task.sh
+then
+  note "GATE PASS $(stamp)"
+  exit 0
+fi
+note "GATE STOPPED $(stamp)"
+exit 1
diff --git a/agents/researcher/work/queue-71/r1/mutants.py b/agents/researcher/work/queue-71/r1/mutants.py
new file mode 100644
index 000000000..1c39a6f45
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/mutants.py
@@ -0,0 +1,95 @@
+#!/usr/bin/env python3
+"""Row 71 round 1 mutants, run from the worktree root after gate.sh passes.
+
+Each mutant is one edit to packages/tasks/src/sync.mjs, applied alone, followed by
+the tasks suite. The original bytes are held in memory and written back in a
+finally block, and the sha256 is checked after the last restore.
+A mutant is killed when the suite exits nonzero. A 124 exit is a timeout and is
+reported as such, not as a kill.
+"""
+import datetime
+import hashlib
+import os
+import re
+import subprocess
+import sys
+
+SYNC = 'packages/tasks/src/sync.mjs'
+P = 'agents/researcher/work/queue-71/r1'
+OUT = P + '/out'
+ORIG_SHA = '35e79dc2ac56af8047bef11a5aaf5f1d37ce3e11513f9968454053068f59ebcf'
+LINE = '  return r.status === 403 || r.status === 404;'
+MUTANTS = [
+    # Filbert N3: the confirm read treats any 4xx as "left".
+    ('r3-confirm-4xx', LINE, '  return r.status >= 400 && r.status < 500;'),
+    # Dewey: the confirm read's failed call (status 0) counts as "left".
+    ('r3-confirm-error-on-throw', LINE, '  return r.status === 403 || r.status === 404 || r.status === 0;'),
+]
+
+
+def stamp():
+    return datetime.datetime.now(datetime.timezone.utc).strftime('%Y-%m-%dT%H:%M:%SZ')
+
+
+def sha(path):
+    with open(path, 'rb') as f:
+        return hashlib.sha256(f.read()).hexdigest()
+
+
+def write(path, data):
+    with open(path, 'wb') as f:
+        f.write(data)
+
+
+def run_suite(name):
+    log = f'{OUT}/mutant-{name}.log'
+    cmd = ['env', 'DOCKER_HOST=unix:///nonexistent.sock', 'node', '--test', '--test-reporter=tap', 'packages/tasks/tests/*.test.mjs']
+    rc = subprocess.run(['python3', P + '/run_group.py', '900', log, os.getcwd(), '--'] + cmd).returncode
+    with open(log, encoding='utf-8', errors='replace') as f:
+        text = f.read()
+    failing = []
+    for m in re.finditer(r'^\s*not ok \d+ - (.+)$', text, re.M):
+        name_ = m.group(1).strip()
+        if name_ not in failing:
+            failing.append(name_)
+    counts = ' '.join(re.findall(r'^# (?:tests|pass|fail) \d+', text, re.M))
+    return rc, failing, counts
+
+
+def main():
+    if not os.path.isfile(SYNC):
+        sys.exit('run from the worktree root')
+    os.makedirs(OUT, exist_ok=True)
+    original = open(SYNC, 'rb').read()
+    if hashlib.sha256(original).hexdigest() != ORIG_SHA:
+        sys.exit('sync.mjs is not the candidate copy; refusing to mutate it')
+    text = original.decode('utf-8')
+    summary = []
+    try:
+        for name, old, new in MUTANTS:
+            if text.count(old + '\n') != 1:
+                sys.exit(f'{name}: target line must occur exactly once')
+            mutated = text.replace(old + '\n', new + '\n').encode('utf-8')
+            write(SYNC, mutated)
+            t0 = stamp()
+            rc, failing, counts = run_suite(name)
+            t1 = stamp()
+            write(SYNC, original)
+            if sha(SYNC) != ORIG_SHA:
+                sys.exit(f'{name}: restore did not reproduce the original sha256')
+            status = 'killed' if rc not in (0, 124) else ('timeout' if rc == 124 else 'SURVIVED')
+            summary.append(f'{name} {status} rc={rc} start={t0} end={t1} {counts}')
+            for n in failing:
+                summary.append(f'    not ok: {n}')
+    finally:
+        write(SYNC, original)
+    final = sha(SYNC)
+    summary.append(f'restored {SYNC} sha256 {final} {"OK" if final == ORIG_SHA else "MISMATCH"}')
+    with open(OUT + '/mutants-summary.txt', 'w', encoding='utf-8') as f:
+        f.write('\n'.join(summary) + '\n')
+    print('\n'.join(summary))
+    sys.exit(0 if final == ORIG_SHA else 1)
+
+
+if __name__ == '__main__':
+    main()
diff --git a/agents/researcher/work/queue-71/r1/out/chain-rc.txt b/agents/researcher/work/queue-71/r1/out/chain-rc.txt
new file mode 100644
index 000000000..474b5ae54
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/chain-rc.txt
@@ -0,0 +1,2 @@
+chain rc=0
+Tue Oct 13 02:48:06 PM UTC 2026
diff --git a/agents/researcher/work/queue-71/r1/out/chain-start.txt b/agents/researcher/work/queue-71/r1/out/chain-start.txt
new file mode 100644
index 000000000..f811bea9c
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/chain-start.txt
@@ -0,0 +1 @@
+Tue Oct 13 02:46:36 PM UTC 2026
diff --git a/agents/researcher/work/queue-71/r1/out/gate-stdout.txt b/agents/researcher/work/queue-71/r1/out/gate-stdout.txt
new file mode 100644
index 000000000..e69de29bb
diff --git a/agents/researcher/work/queue-71/r1/out/gate-summary.txt b/agents/researcher/work/queue-71/r1/out/gate-summary.txt
new file mode 100644
index 000000000..f2d0391c8
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/gate-summary.txt
@@ -0,0 +1,13 @@
+gate start 2026-10-13T14:46:36Z base ac6b56ee55278077d61f39e6ca42513725427154
+node-tasks rc=0 start=2026-10-13T14:46:36Z end=2026-10-13T14:46:37Z # tests 89 # pass 89 # fail 0 # cancelled 0 # skipped 0 # todo 0 
+node-bus rc=0 start=2026-10-13T14:46:37Z end=2026-10-13T14:46:38Z # tests 115 # pass 115 # fail 0 # cancelled 0 # skipped 0 # todo 0 
+test-auth rc=0 start=2026-10-13T14:46:38Z end=2026-10-13T14:46:38Z 
+test-conductor rc=0 start=2026-10-13T14:46:38Z end=2026-10-13T14:46:44Z 
+test-config rc=0 start=2026-10-13T14:46:44Z end=2026-10-13T14:46:46Z 
+test-discord rc=0 start=2026-10-13T14:46:46Z end=2026-10-13T14:46:53Z 
+test-extension-package rc=0 start=2026-10-13T14:46:53Z end=2026-10-13T14:46:54Z 
+test-foundation rc=0 start=2026-10-13T14:46:54Z end=2026-10-13T14:47:32Z 
+test-queue rc=0 start=2026-10-13T14:47:32Z end=2026-10-13T14:47:52Z 
+test-release rc=0 start=2026-10-13T14:47:52Z end=2026-10-13T14:48:02Z 
+test-task rc=0 start=2026-10-13T14:48:02Z end=2026-10-13T14:48:04Z 
+GATE PASS 2026-10-13T14:48:04Z
diff --git a/agents/researcher/work/queue-71/r1/out/manifest-check.txt b/agents/researcher/work/queue-71/r1/out/manifest-check.txt
new file mode 100644
index 000000000..644ff2d8c
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/manifest-check.txt
@@ -0,0 +1,2 @@
+packages/tasks/src/fake.mjs: OK
+packages/tasks/tests/tree.test.mjs: OK
diff --git a/agents/researcher/work/queue-71/r1/out/mutant-r3-confirm-4xx.log b/agents/researcher/work/queue-71/r1/out/mutant-r3-confirm-4xx.log
new file mode 100644
index 000000000..7a24e265e
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/mutant-r3-confirm-4xx.log
@@ -0,0 +1,556 @@
+TAP version 13
+# Subtest: the boot config is checked before anything starts
+ok 1 - the boot config is checked before anything starts
+  ---
+  duration_ms: 14.985128
+  type: 'test'
+  ...
+# Subtest: a business with no tracker entry refuses task verbs
+ok 2 - a business with no tracker entry refuses task verbs
+  ---
+  duration_ms: 14.714455
+  type: 'test'
+  ...
+# Subtest: credential.expiring and .expired are recorded once per instance
+ok 3 - credential.expiring and .expired are recorded once per instance
+  ---
+  duration_ms: 27.882089
+  type: 'test'
+  ...
+# Subtest: a token file that changes on disk records credential.changed
+ok 4 - a token file that changes on disk records credential.changed
+  ---
+  duration_ms: 13.348847
+  type: 'test'
+  ...
+# Subtest: autostart polls, reconciles and retries a startup the tracker was down for
+ok 5 - autostart polls, reconciles and retries a startup the tracker was down for
+  ---
+  duration_ms: 36.23427
+  type: 'test'
+  ...
+# Subtest: a refusal a restart must clear is not retried by the poll
+ok 6 - a refusal a restart must clear is not retried by the poll
+  ---
+  duration_ms: 10.973461
+  type: 'test'
+  ...
+# Subtest: a poll that fires while two are queued is dropped
+ok 7 - a poll that fires while two are queued is dropped
+  ---
+  duration_ms: 14.054849
+  type: 'test'
+  ...
+# Subtest: close waits for a running verb and refuses one that has not started
+ok 8 - close waits for a running verb and refuses one that has not started
+  ---
+  duration_ms: 44.044153
+  type: 'test'
+  ...
+# Subtest: the bundled Vikunja is the pinned upstream image the runbook names
+ok 9 - the bundled Vikunja is the pinned upstream image the runbook names
+  ---
+  duration_ms: 0.711278
+  type: 'test'
+  ...
+# Subtest: every published port is on 127.0.0.1, and no secret is in the file
+ok 10 - every published port is on 127.0.0.1, and no secret is in the file
+  ---
+  duration_ms: 0.258663
+  type: 'test'
+  ...
+# Subtest: the committed table loads frozen; its verbs are the broker task verbs
+ok 11 - the committed table loads frozen; its verbs are the broker task verbs
+  ---
+  duration_ms: 1.669028
+  type: 'test'
+  ...
+# Subtest: the schema refuses any other shape
+ok 12 - the schema refuses any other shape
+  ---
+  duration_ms: 3.318884
+  type: 'test'
+  ...
+# Subtest: each verb refuses, asks for a decision and writes as the table and the role files say
+ok 13 - each verb refuses, asks for a decision and writes as the table and the role files say
+  ---
+  duration_ms: 155.091502
+  type: 'test'
+  ...
+# Subtest: the list holds the PRD's requirement ids, in its order, at its version
+ok 14 - the list holds the PRD's requirement ids, in its order, at its version
+  ---
+  duration_ms: 5.451833
+  type: 'test'
+  ...
+# Subtest: the committed list loads frozen
+ok 15 - the committed list loads frozen
+  ---
+  duration_ms: 0.657564
+  type: 'test'
+  ...
+# Subtest: the schema refuses any other shape
+ok 16 - the schema refuses any other shape
+  ---
+  duration_ms: 1.923796
+  type: 'test'
+  ...
+# Subtest: the fake answers each route with the statuses and shapes Vikunja v2.7.0 sent
+ok 17 - the fake answers each route with the statuses and shapes Vikunja v2.7.0 sent
+  ---
+  duration_ms: 91.483598
+  type: 'test'
+  ...
+# Subtest: the project list holds recorded project objects in the recorded list envelope
+ok 18 - the project list holds recorded project objects in the recorded list envelope
+  ---
+  duration_ms: 9.898596
+  type: 'test'
+  ...
+# Subtest: the recorded task bodies pass the checks S3 applies to every read
+ok 19 - the recorded task bodies pass the checks S3 applies to every read
+  ---
+  duration_ms: 0.43417
+  type: 'test'
+  ...
+# Subtest: the client works against the fake over real HTTP with the platform fetch
+ok 20 - the client works against the fake over real HTTP with the platform fetch
+  ---
+  duration_ms: 99.305719
+  type: 'test'
+  ...
+# Subtest: a correct install starts, and the first reconcile records tasks that already exist
+ok 21 - a correct install starts, and the first reconcile records tasks that already exist
+  ---
+  duration_ms: 26.885137
+  type: 'test'
+  ...
+# Subtest: verbs refuse while a business is starting and after startup refused it
+ok 22 - verbs refuse while a business is starting and after startup refused it
+  ---
+  duration_ms: 15.866064
+  type: 'test'
+  ...
+# Subtest: startup refuses a token that can do more than its role needs
+ok 23 - startup refuses a token that can do more than its role needs
+  ---
+  duration_ms: 39.459007
+  type: 'test'
+  ...
+# Subtest: startup refuses an unsupported version and flags an untested one
+ok 24 - startup refuses an unsupported version and flags an untested one
+  ---
+  duration_ms: 29.725668
+  type: 'test'
+  ...
+# Subtest: startup refuses a board that the runbook did not install
+ok 25 - startup refuses a board that the runbook did not install
+  ---
+  duration_ms: 40.235276
+  type: 'test'
+  ...
+# Subtest: startup refuses a project the sync bot cannot read
+ok 26 - startup refuses a project the sync bot cannot read
+  ---
+  duration_ms: 8.170918
+  type: 'test'
+  ...
+# Subtest: startup refuses a configured label the pm bot cannot see
+ok 27 - startup refuses a configured label the pm bot cannot see
+  ---
+  duration_ms: 9.844785
+  type: 'test'
+  ...
+# Subtest: startup refuses an expired credential and a missing sync credential
+ok 28 - startup refuses an expired credential and a missing sync credential
+  ---
+  duration_ms: 16.9923
+  type: 'test'
+  ...
+# Subtest: an unreachable tracker refuses with tracker-unavailable
+ok 29 - an unreachable tracker refuses with tracker-unavailable
+  ---
+  duration_ms: 6.645521
+  type: 'test'
+  ...
+# Subtest: an edit in the UI is recorded once, with the fields that changed
+ok 30 - an edit in the UI is recorded once, with the fields that changed
+  ---
+  duration_ms: 77.08308
+  type: 'test'
+  ...
+# Subtest: a move between open buckets is seen on the board, though updated does not change
+ok 31 - a move between open buckets is seen on the board, though updated does not change
+  ---
+  duration_ms: 39.81042
+  type: 'test'
+  ...
+# Subtest: a person's comment is counted and a bot's is not
+ok 32 - a person's comment is counted and a bot's is not
+  ---
+  duration_ms: 73.111456
+  type: 'test'
+  ...
+# Subtest: the hourly reconcile catches a comment through comment_count
+ok 33 - the hourly reconcile catches a comment through comment_count
+  ---
+  duration_ms: 76.173236
+  type: 'test'
+  ...
+# Subtest: a task closed in the UI leaves the open view with its done bucket
+ok 34 - a task closed in the UI leaves the open view with its done bucket
+  ---
+  duration_ms: 113.942375
+  type: 'test'
+  ...
+# Subtest: a task that leaves the board is recorded as deleted, moved or out of reach
+ok 35 - a task that leaves the board is recorded as deleted, moved or out of reach
+  ---
+  duration_ms: 64.372593
+  type: 'test'
+  ...
+# Subtest: a poll that read before a verb wrote does not overwrite the verb
+ok 36 - a poll that read before a verb wrote does not overwrite the verb
+  ---
+  duration_ms: 35.608068
+  type: 'test'
+  ...
+# Subtest: a tracker fault during a tick is reported and the next tick catches up
+ok 37 - a tracker fault during a tick is reported and the next tick catches up
+  ---
+  duration_ms: 35.704917
+  type: 'test'
+  ...
+# Subtest: a malformed answer refuses the tick with tracker-shape
+ok 38 - a malformed answer refuses the tick with tracker-shape
+  ---
+  duration_ms: 13.652804
+  type: 'test'
+  ...
+# Subtest: no token value reaches the database, the log or a refusal
+ok 39 - no token value reaches the database, the log or a refusal
+  ---
+  duration_ms: 43.73452
+  type: 'test'
+  ...
+# Subtest: the first look at a task counts only comments inside the window
+ok 40 - the first look at a task counts only comments inside the window
+  ---
+  duration_ms: 33.293957
+  type: 'test'
+  ...
+# Subtest: with descendants off the adapter reads no project list and no child, as in v0
+ok 41 - with descendants off the adapter reads no project list and no child, as in v0
+  ---
+  duration_ms: 46.419107
+  type: 'test'
+  ...
+# Subtest: a three-level tree is read per project, and a project outside it is not
+ok 42 - a three-level tree is read per project, and a project outside it is not
+  ---
+  duration_ms: 19.063862
+  type: 'test'
+  ...
+# Subtest: a watch-only child keeps the business ready, and a failed child read refuses as transient
+ok 43 - a watch-only child keeps the business ready, and a failed child read refuses as transient
+  ---
+  duration_ms: 24.274217
+  type: 'test'
+  ...
+# Subtest: a root without the installed buckets refuses with descendants on
+ok 44 - a root without the installed buckets refuses with descendants on
+  ---
+  duration_ms: 10.156206
+  type: 'test'
+  ...
+# Subtest: the project list refuses without read_all, and skips archived projects and saved filters
+ok 45 - the project list refuses without read_all, and skips archived projects and saved filters
+  ---
+  duration_ms: 22.653056
+  type: 'test'
+  ...
+# Subtest: descendants must be true or false
+ok 46 - descendants must be true or false
+  ---
+  duration_ms: 11.059931
+  type: 'test'
+  ...
+# Subtest: a move inside the tree is a moved tombstone plus a snapshot under the new ref
+ok 47 - a move inside the tree is a moved tombstone plus a snapshot under the new ref
+  ---
+  duration_ms: 65.339785
+  type: 'test'
+  ...
+# Subtest: a child made after startup joins the tree at the next reconcile
+ok 48 - a child made after startup joins the tree at the next reconcile
+  ---
+  duration_ms: 52.570713
+  type: 'test'
+  ...
+# Subtest: a child that leaves the tree has its tasks recorded missing, and the ledger keeps their history
+ok 49 - a child that leaves the tree has its tasks recorded missing, and the ledger keeps their history
+  ---
+  duration_ms: 58.458871
+  type: 'test'
+  ...
+# Subtest: a child deleted or hidden between reconciles drops out at the next tick, and the root still refuses
+ok 50 - a child deleted or hidden between reconciles drops out at the next tick, and the root still refuses
+  ---
+  duration_ms: 82.229086
+  type: 'test'
+  ...
+# Subtest: a reconcile that fails keeps the previous tree
+ok 51 - a reconcile that fails keeps the previous tree
+  ---
+  duration_ms: 32.227599
+  type: 'test'
+  ...
+# Subtest: a closed task the cursor missed takes its own project's done bucket
+ok 52 - a closed task the cursor missed takes its own project's done bucket
+  ---
+  duration_ms: 32.350006
+  type: 'test'
+  ...
+# Subtest: closed tasks read at startup take their own project's done bucket
+ok 53 - closed tasks read at startup take their own project's done bucket
+  ---
+  duration_ms: 8.718222
+  type: 'test'
+  ...
+# Subtest: a closed task the cursor returns takes its own project's done bucket
+ok 54 - a closed task the cursor returns takes its own project's done bucket
+  ---
+  duration_ms: 32.245047
+  type: 'test'
+  ...
+# Subtest: a task moved out of a child that leaves the tree at the same time is moved, not no-access
+ok 55 - a task moved out of a child that leaves the tree at the same time is moved, not no-access
+  ---
+  duration_ms: 31.487855
+  type: 'test'
+  ...
+# Subtest: a task in a bucket outside the five keeps its raw bucket id
+ok 56 - a task in a bucket outside the five keeps its raw bucket id
+  ---
+  duration_ms: 30.317739
+  type: 'test'
+  ...
+# Subtest: bucket 0 on a watch-only task never becomes a bucket or a state
+ok 57 - bucket 0 on a watch-only task never becomes a bucket or a state
+  ---
+  duration_ms: 76.762699
+  type: 'test'
+  ...
+# Subtest: a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+not ok 58 - a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+  ---
+  duration_ms: 34.653246
+  type: 'test'
+  location: '/mnt/storage/src/mosaic-stack/.claude/worktrees/agent-a44468a48c485520e/packages/tasks/tests/tree.test.mjs:425:1'
+  failureType: 'testCodeFailure'
+  error: 'Missing expected rejection.'
+  code: 'ERR_ASSERTION'
+  name: 'AssertionError'
+  expected:
+    code: 'tracker-not-found'
+  operator: 'rejects'
+  stack: |-
+    async TestContext.<anonymous> (file:///mnt/storage/src/mosaic-stack/.claude/worktrees/agent-a44468a48c485520e/packages/tasks/tests/tree.test.mjs:465:3)
+    async Test.run (node:internal/test_runner/test:1409:7)
+    async Test.processPendingSubtests (node:internal/test_runner/test:974:7)
+  ...
+# Subtest: a child dropped at a tick comes back at the next reconcile once the bot reads it again
+ok 59 - a child dropped at a tick comes back at the next reconcile once the bot reads it again
+  ---
+  duration_ms: 39.468458
+  type: 'test'
+  ...
+# Subtest: task.create needs a recorded human request and a requirement id
+ok 60 - task.create needs a recorded human request and a requirement id
+  ---
+  duration_ms: 37.76007
+  type: 'test'
+  ...
+# Subtest: task.create takes only an approved requirement id
+ok 61 - task.create takes only an approved requirement id
+  ---
+  duration_ms: 24.103231
+  type: 'test'
+  ...
+# Subtest: task.create cites the request a received message carries, not the message's id
+ok 62 - task.create cites the request a received message carries, not the message's id
+  ---
+  duration_ms: 19.530799
+  type: 'test'
+  ...
+# Subtest: only labels named in the business file can be written
+ok 63 - only labels named in the business file can be written
+  ---
+  duration_ms: 25.956217
+  type: 'test'
+  ...
+# Subtest: task.schedule sets and clears a due date and relations
+ok 64 - task.schedule sets and clears a due date and relations
+  ---
+  duration_ms: 38.314616
+  type: 'test'
+  ...
+# Subtest: assign and reassign move the role bots and record task.assigned
+ok 65 - assign and reassign move the role bots and record task.assigned
+  ---
+  duration_ms: 55.376954
+  type: 'test'
+  ...
+# Subtest: task.update.assigned is for the assignee and records task.state
+ok 66 - task.update.assigned is for the assignee and records task.state
+  ---
+  duration_ms: 43.392796
+  type: 'test'
+  ...
+# Subtest: a wrong expected digest records task.conflict and writes nothing
+ok 67 - a wrong expected digest records task.conflict and writes nothing
+  ---
+  duration_ms: 19.114486
+  type: 'test'
+  ...
+# Subtest: a cross-role verb needs a resolved decision, used once
+ok 68 - a cross-role verb needs a resolved decision, used once
+  ---
+  duration_ms: 20.735188
+  type: 'test'
+  ...
+# Subtest: task.close needs a verdict; after it every verb refuses with task-done
+ok 69 - task.close needs a verdict; after it every verb refuses with task-done
+  ---
+  duration_ms: 21.0694
+  type: 'test'
+  ...
+# Subtest: a lost answer is settled by a re-read and never retried
+ok 70 - a lost answer is settled by a re-read and never retried
+  ---
+  duration_ms: 25.470917
+  type: 'test'
+  ...
+# Subtest: a create whose answer is lost is reported uncertain, and the poll finds the task
+ok 71 - a create whose answer is lost is reported uncertain, and the poll finds the task
+  ---
+  duration_ms: 34.598261
+  type: 'test'
+  ...
+# Subtest: a task the sync bot cannot read refuses and records nothing
+ok 72 - a task the sync bot cannot read refuses and records nothing
+  ---
+  duration_ms: 17.107712
+  type: 'test'
+  ...
+# Subtest: verbs and polls for one business run one at a time
+ok 73 - verbs and polls for one business run one at a time
+  ---
+  duration_ms: 87.088201
+  type: 'test'
+  ...
+# Subtest: a due date with milliseconds is written to the second
+ok 74 - a due date with milliseconds is written to the second
+  ---
+  duration_ms: 62.660034
+  type: 'test'
+  ...
+# Subtest: every write landed and the final read failed: the verb succeeds and records what it wrote
+ok 75 - every write landed and the final read failed: the verb succeeds and records what it wrote
+  ---
+  duration_ms: 35.866185
+  type: 'test'
+  ...
+# Subtest: some writes landed and the final read failed: write-uncertain, and nothing is recorded
+ok 76 - some writes landed and the final read failed: write-uncertain, and nothing is recorded
+  ---
+  duration_ms: 13.780501
+  type: 'test'
+  ...
+# Subtest: a create whose final read fails succeeds and records task.created
+ok 77 - a create whose final read fails succeeds and records task.created
+  ---
+  duration_ms: 31.439164
+  type: 'test'
+  ...
+# Subtest: an edit between the last write and the final read shows as external on the next poll
+ok 78 - an edit between the last write and the final read shows as external on the next poll
+  ---
+  duration_ms: 34.514959
+  type: 'test'
+  ...
+# Subtest: task.created is recorded when a later label write fails
+ok 79 - task.created is recorded when a later label write fails
+  ---
+  duration_ms: 15.127095
+  type: 'test'
+  ...
+# Subtest: every verb acts in an installed child, on the child's own board
+ok 80 - every verb acts in an installed child, on the child's own board
+  ---
+  duration_ms: 54.093416
+  type: 'test'
+  ...
+# Subtest: task.create refuses a project outside the tree, a watch-only one and one that is not an id
+ok 81 - task.create refuses a project outside the tree, a watch-only one and one that is not an id
+  ---
+  duration_ms: 28.174679
+  type: 'test'
+  ...
+# Subtest: with descendants off, task.create takes only the root project
+ok 82 - with descendants off, task.create takes only the root project
+  ---
+  duration_ms: 14.206089
+  type: 'test'
+  ...
+# Subtest: a verb on a watch-only task, or a relation to one, refuses with project-not-installed
+ok 83 - a verb on a watch-only task, or a relation to one, refuses with project-not-installed
+  ---
+  duration_ms: 19.361082
+  type: 'test'
+  ...
+# Subtest: a related task is read before authority: in another project than its ref names, task-moved; missing, task-not-found
+ok 84 - a related task is read before authority: in another project than its ref names, task-moved; missing, task-not-found
+  ---
+  duration_ms: 24.895956
+  type: 'test'
+  ...
+# Subtest: a task moved to another project of the tree refuses its old ref with task-moved and takes its new one
+ok 85 - a task moved to another project of the tree refuses its old ref with task-moved and takes its new one
+  ---
+  duration_ms: 17.406278
+  type: 'test'
+  ...
+# Subtest: a verb and a create in the child record the final read of the child's board
+ok 86 - a verb and a create in the child record the final read of the child's board
+  ---
+  duration_ms: 15.833158
+  type: 'test'
+  ...
+# Subtest: a create in the child whose final read fails records the child's todo bucket
+ok 87 - a create in the child whose final read fails records the child's todo bucket
+  ---
+  duration_ms: 35.839631
+  type: 'test'
+  ...
+# Subtest: a lost answer in the child is settled by a re-read of the child's board
+ok 88 - a lost answer in the child is settled by a re-read of the child's board
+  ---
+  duration_ms: 23.642965
+  type: 'test'
+  ...
+# Subtest: every related task is read before authority, not only the first
+ok 89 - every related task is read before authority, not only the first
+  ---
+  duration_ms: 18.46152
+  type: 'test'
+  ...
+1..89
+# tests 89
+# suites 0
+# pass 88
+# fail 1
+# cancelled 0
+# skipped 0
+# todo 0
+# duration_ms 1018.079754
diff --git a/agents/researcher/work/queue-71/r1/out/mutant-r3-confirm-error-on-throw.log b/agents/researcher/work/queue-71/r1/out/mutant-r3-confirm-error-on-throw.log
new file mode 100644
index 000000000..4df4312db
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/mutant-r3-confirm-error-on-throw.log
@@ -0,0 +1,556 @@
+TAP version 13
+# Subtest: the boot config is checked before anything starts
+ok 1 - the boot config is checked before anything starts
+  ---
+  duration_ms: 30.211993
+  type: 'test'
+  ...
+# Subtest: a business with no tracker entry refuses task verbs
+ok 2 - a business with no tracker entry refuses task verbs
+  ---
+  duration_ms: 13.221941
+  type: 'test'
+  ...
+# Subtest: credential.expiring and .expired are recorded once per instance
+ok 3 - credential.expiring and .expired are recorded once per instance
+  ---
+  duration_ms: 30.026771
+  type: 'test'
+  ...
+# Subtest: a token file that changes on disk records credential.changed
+ok 4 - a token file that changes on disk records credential.changed
+  ---
+  duration_ms: 13.716724
+  type: 'test'
+  ...
+# Subtest: autostart polls, reconciles and retries a startup the tracker was down for
+ok 5 - autostart polls, reconciles and retries a startup the tracker was down for
+  ---
+  duration_ms: 37.157349
+  type: 'test'
+  ...
+# Subtest: a refusal a restart must clear is not retried by the poll
+ok 6 - a refusal a restart must clear is not retried by the poll
+  ---
+  duration_ms: 12.484131
+  type: 'test'
+  ...
+# Subtest: a poll that fires while two are queued is dropped
+ok 7 - a poll that fires while two are queued is dropped
+  ---
+  duration_ms: 14.564878
+  type: 'test'
+  ...
+# Subtest: close waits for a running verb and refuses one that has not started
+ok 8 - close waits for a running verb and refuses one that has not started
+  ---
+  duration_ms: 43.010075
+  type: 'test'
+  ...
+# Subtest: the bundled Vikunja is the pinned upstream image the runbook names
+ok 9 - the bundled Vikunja is the pinned upstream image the runbook names
+  ---
+  duration_ms: 0.782649
+  type: 'test'
+  ...
+# Subtest: every published port is on 127.0.0.1, and no secret is in the file
+ok 10 - every published port is on 127.0.0.1, and no secret is in the file
+  ---
+  duration_ms: 0.284211
+  type: 'test'
+  ...
+# Subtest: the committed table loads frozen; its verbs are the broker task verbs
+ok 11 - the committed table loads frozen; its verbs are the broker task verbs
+  ---
+  duration_ms: 1.550886
+  type: 'test'
+  ...
+# Subtest: the schema refuses any other shape
+ok 12 - the schema refuses any other shape
+  ---
+  duration_ms: 3.798377
+  type: 'test'
+  ...
+# Subtest: each verb refuses, asks for a decision and writes as the table and the role files say
+ok 13 - each verb refuses, asks for a decision and writes as the table and the role files say
+  ---
+  duration_ms: 161.456593
+  type: 'test'
+  ...
+# Subtest: the list holds the PRD's requirement ids, in its order, at its version
+ok 14 - the list holds the PRD's requirement ids, in its order, at its version
+  ---
+  duration_ms: 5.001933
+  type: 'test'
+  ...
+# Subtest: the committed list loads frozen
+ok 15 - the committed list loads frozen
+  ---
+  duration_ms: 0.513652
+  type: 'test'
+  ...
+# Subtest: the schema refuses any other shape
+ok 16 - the schema refuses any other shape
+  ---
+  duration_ms: 2.118521
+  type: 'test'
+  ...
+# Subtest: the fake answers each route with the statuses and shapes Vikunja v2.7.0 sent
+ok 17 - the fake answers each route with the statuses and shapes Vikunja v2.7.0 sent
+  ---
+  duration_ms: 100.558772
+  type: 'test'
+  ...
+# Subtest: the project list holds recorded project objects in the recorded list envelope
+ok 18 - the project list holds recorded project objects in the recorded list envelope
+  ---
+  duration_ms: 9.625038
+  type: 'test'
+  ...
+# Subtest: the recorded task bodies pass the checks S3 applies to every read
+ok 19 - the recorded task bodies pass the checks S3 applies to every read
+  ---
+  duration_ms: 0.510864
+  type: 'test'
+  ...
+# Subtest: the client works against the fake over real HTTP with the platform fetch
+ok 20 - the client works against the fake over real HTTP with the platform fetch
+  ---
+  duration_ms: 99.450496
+  type: 'test'
+  ...
+# Subtest: a correct install starts, and the first reconcile records tasks that already exist
+ok 21 - a correct install starts, and the first reconcile records tasks that already exist
+  ---
+  duration_ms: 35.875385
+  type: 'test'
+  ...
+# Subtest: verbs refuse while a business is starting and after startup refused it
+ok 22 - verbs refuse while a business is starting and after startup refused it
+  ---
+  duration_ms: 15.070202
+  type: 'test'
+  ...
+# Subtest: startup refuses a token that can do more than its role needs
+ok 23 - startup refuses a token that can do more than its role needs
+  ---
+  duration_ms: 44.434924
+  type: 'test'
+  ...
+# Subtest: startup refuses an unsupported version and flags an untested one
+ok 24 - startup refuses an unsupported version and flags an untested one
+  ---
+  duration_ms: 30.974458
+  type: 'test'
+  ...
+# Subtest: startup refuses a board that the runbook did not install
+ok 25 - startup refuses a board that the runbook did not install
+  ---
+  duration_ms: 39.732346
+  type: 'test'
+  ...
+# Subtest: startup refuses a project the sync bot cannot read
+ok 26 - startup refuses a project the sync bot cannot read
+  ---
+  duration_ms: 6.867169
+  type: 'test'
+  ...
+# Subtest: startup refuses a configured label the pm bot cannot see
+ok 27 - startup refuses a configured label the pm bot cannot see
+  ---
+  duration_ms: 8.904923
+  type: 'test'
+  ...
+# Subtest: startup refuses an expired credential and a missing sync credential
+ok 28 - startup refuses an expired credential and a missing sync credential
+  ---
+  duration_ms: 13.255291
+  type: 'test'
+  ...
+# Subtest: an unreachable tracker refuses with tracker-unavailable
+ok 29 - an unreachable tracker refuses with tracker-unavailable
+  ---
+  duration_ms: 6.283231
+  type: 'test'
+  ...
+# Subtest: an edit in the UI is recorded once, with the fields that changed
+ok 30 - an edit in the UI is recorded once, with the fields that changed
+  ---
+  duration_ms: 79.358718
+  type: 'test'
+  ...
+# Subtest: a move between open buckets is seen on the board, though updated does not change
+ok 31 - a move between open buckets is seen on the board, though updated does not change
+  ---
+  duration_ms: 41.16021
+  type: 'test'
+  ...
+# Subtest: a person's comment is counted and a bot's is not
+ok 32 - a person's comment is counted and a bot's is not
+  ---
+  duration_ms: 72.985532
+  type: 'test'
+  ...
+# Subtest: the hourly reconcile catches a comment through comment_count
+ok 33 - the hourly reconcile catches a comment through comment_count
+  ---
+  duration_ms: 76.813791
+  type: 'test'
+  ...
+# Subtest: a task closed in the UI leaves the open view with its done bucket
+ok 34 - a task closed in the UI leaves the open view with its done bucket
+  ---
+  duration_ms: 113.610013
+  type: 'test'
+  ...
+# Subtest: a task that leaves the board is recorded as deleted, moved or out of reach
+ok 35 - a task that leaves the board is recorded as deleted, moved or out of reach
+  ---
+  duration_ms: 65.693133
+  type: 'test'
+  ...
+# Subtest: a poll that read before a verb wrote does not overwrite the verb
+ok 36 - a poll that read before a verb wrote does not overwrite the verb
+  ---
+  duration_ms: 34.519446
+  type: 'test'
+  ...
+# Subtest: a tracker fault during a tick is reported and the next tick catches up
+ok 37 - a tracker fault during a tick is reported and the next tick catches up
+  ---
+  duration_ms: 35.540654
+  type: 'test'
+  ...
+# Subtest: a malformed answer refuses the tick with tracker-shape
+ok 38 - a malformed answer refuses the tick with tracker-shape
+  ---
+  duration_ms: 11.963452
+  type: 'test'
+  ...
+# Subtest: no token value reaches the database, the log or a refusal
+ok 39 - no token value reaches the database, the log or a refusal
+  ---
+  duration_ms: 46.592746
+  type: 'test'
+  ...
+# Subtest: the first look at a task counts only comments inside the window
+ok 40 - the first look at a task counts only comments inside the window
+  ---
+  duration_ms: 34.905182
+  type: 'test'
+  ...
+# Subtest: with descendants off the adapter reads no project list and no child, as in v0
+ok 41 - with descendants off the adapter reads no project list and no child, as in v0
+  ---
+  duration_ms: 45.272597
+  type: 'test'
+  ...
+# Subtest: a three-level tree is read per project, and a project outside it is not
+ok 42 - a three-level tree is read per project, and a project outside it is not
+  ---
+  duration_ms: 21.475483
+  type: 'test'
+  ...
+# Subtest: a watch-only child keeps the business ready, and a failed child read refuses as transient
+ok 43 - a watch-only child keeps the business ready, and a failed child read refuses as transient
+  ---
+  duration_ms: 26.92962
+  type: 'test'
+  ...
+# Subtest: a root without the installed buckets refuses with descendants on
+ok 44 - a root without the installed buckets refuses with descendants on
+  ---
+  duration_ms: 10.679158
+  type: 'test'
+  ...
+# Subtest: the project list refuses without read_all, and skips archived projects and saved filters
+ok 45 - the project list refuses without read_all, and skips archived projects and saved filters
+  ---
+  duration_ms: 23.327317
+  type: 'test'
+  ...
+# Subtest: descendants must be true or false
+ok 46 - descendants must be true or false
+  ---
+  duration_ms: 9.00679
+  type: 'test'
+  ...
+# Subtest: a move inside the tree is a moved tombstone plus a snapshot under the new ref
+ok 47 - a move inside the tree is a moved tombstone plus a snapshot under the new ref
+  ---
+  duration_ms: 63.81362
+  type: 'test'
+  ...
+# Subtest: a child made after startup joins the tree at the next reconcile
+ok 48 - a child made after startup joins the tree at the next reconcile
+  ---
+  duration_ms: 53.761186
+  type: 'test'
+  ...
+# Subtest: a child that leaves the tree has its tasks recorded missing, and the ledger keeps their history
+ok 49 - a child that leaves the tree has its tasks recorded missing, and the ledger keeps their history
+  ---
+  duration_ms: 58.663188
+  type: 'test'
+  ...
+# Subtest: a child deleted or hidden between reconciles drops out at the next tick, and the root still refuses
+ok 50 - a child deleted or hidden between reconciles drops out at the next tick, and the root still refuses
+  ---
+  duration_ms: 81.807043
+  type: 'test'
+  ...
+# Subtest: a reconcile that fails keeps the previous tree
+ok 51 - a reconcile that fails keeps the previous tree
+  ---
+  duration_ms: 32.544718
+  type: 'test'
+  ...
+# Subtest: a closed task the cursor missed takes its own project's done bucket
+ok 52 - a closed task the cursor missed takes its own project's done bucket
+  ---
+  duration_ms: 33.196306
+  type: 'test'
+  ...
+# Subtest: closed tasks read at startup take their own project's done bucket
+ok 53 - closed tasks read at startup take their own project's done bucket
+  ---
+  duration_ms: 9.152244
+  type: 'test'
+  ...
+# Subtest: a closed task the cursor returns takes its own project's done bucket
+ok 54 - a closed task the cursor returns takes its own project's done bucket
+  ---
+  duration_ms: 33.025508
+  type: 'test'
+  ...
+# Subtest: a task moved out of a child that leaves the tree at the same time is moved, not no-access
+ok 55 - a task moved out of a child that leaves the tree at the same time is moved, not no-access
+  ---
+  duration_ms: 31.314653
+  type: 'test'
+  ...
+# Subtest: a task in a bucket outside the five keeps its raw bucket id
+ok 56 - a task in a bucket outside the five keeps its raw bucket id
+  ---
+  duration_ms: 31.721671
+  type: 'test'
+  ...
+# Subtest: bucket 0 on a watch-only task never becomes a bucket or a state
+ok 57 - bucket 0 on a watch-only task never becomes a bucket or a state
+  ---
+  duration_ms: 76.658771
+  type: 'test'
+  ...
+# Subtest: a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+not ok 58 - a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+  ---
+  duration_ms: 34.807865
+  type: 'test'
+  location: '/mnt/storage/src/mosaic-stack/.claude/worktrees/agent-a44468a48c485520e/packages/tasks/tests/tree.test.mjs:425:1'
+  failureType: 'testCodeFailure'
+  error: 'Missing expected rejection.'
+  code: 'ERR_ASSERTION'
+  name: 'AssertionError'
+  expected:
+    code: 'tracker-not-found'
+  operator: 'rejects'
+  stack: |-
+    async TestContext.<anonymous> (file:///mnt/storage/src/mosaic-stack/.claude/worktrees/agent-a44468a48c485520e/packages/tasks/tests/tree.test.mjs:472:3)
+    async Test.run (node:internal/test_runner/test:1409:7)
+    async Test.processPendingSubtests (node:internal/test_runner/test:974:7)
+  ...
+# Subtest: a child dropped at a tick comes back at the next reconcile once the bot reads it again
+ok 59 - a child dropped at a tick comes back at the next reconcile once the bot reads it again
+  ---
+  duration_ms: 36.556206
+  type: 'test'
+  ...
+# Subtest: task.create needs a recorded human request and a requirement id
+ok 60 - task.create needs a recorded human request and a requirement id
+  ---
+  duration_ms: 37.477814
+  type: 'test'
+  ...
+# Subtest: task.create takes only an approved requirement id
+ok 61 - task.create takes only an approved requirement id
+  ---
+  duration_ms: 25.776035
+  type: 'test'
+  ...
+# Subtest: task.create cites the request a received message carries, not the message's id
+ok 62 - task.create cites the request a received message carries, not the message's id
+  ---
+  duration_ms: 21.428282
+  type: 'test'
+  ...
+# Subtest: only labels named in the business file can be written
+ok 63 - only labels named in the business file can be written
+  ---
+  duration_ms: 30.604156
+  type: 'test'
+  ...
+# Subtest: task.schedule sets and clears a due date and relations
+ok 64 - task.schedule sets and clears a due date and relations
+  ---
+  duration_ms: 35.696893
+  type: 'test'
+  ...
+# Subtest: assign and reassign move the role bots and record task.assigned
+ok 65 - assign and reassign move the role bots and record task.assigned
+  ---
+  duration_ms: 51.677173
+  type: 'test'
+  ...
+# Subtest: task.update.assigned is for the assignee and records task.state
+ok 66 - task.update.assigned is for the assignee and records task.state
+  ---
+  duration_ms: 44.590124
+  type: 'test'
+  ...
+# Subtest: a wrong expected digest records task.conflict and writes nothing
+ok 67 - a wrong expected digest records task.conflict and writes nothing
+  ---
+  duration_ms: 17.881052
+  type: 'test'
+  ...
+# Subtest: a cross-role verb needs a resolved decision, used once
+ok 68 - a cross-role verb needs a resolved decision, used once
+  ---
+  duration_ms: 20.322939
+  type: 'test'
+  ...
+# Subtest: task.close needs a verdict; after it every verb refuses with task-done
+ok 69 - task.close needs a verdict; after it every verb refuses with task-done
+  ---
+  duration_ms: 21.187137
+  type: 'test'
+  ...
+# Subtest: a lost answer is settled by a re-read and never retried
+ok 70 - a lost answer is settled by a re-read and never retried
+  ---
+  duration_ms: 25.142297
+  type: 'test'
+  ...
+# Subtest: a create whose answer is lost is reported uncertain, and the poll finds the task
+ok 71 - a create whose answer is lost is reported uncertain, and the poll finds the task
+  ---
+  duration_ms: 36.049367
+  type: 'test'
+  ...
+# Subtest: a task the sync bot cannot read refuses and records nothing
+ok 72 - a task the sync bot cannot read refuses and records nothing
+  ---
+  duration_ms: 15.744434
+  type: 'test'
+  ...
+# Subtest: verbs and polls for one business run one at a time
+ok 73 - verbs and polls for one business run one at a time
+  ---
+  duration_ms: 87.455266
+  type: 'test'
+  ...
+# Subtest: a due date with milliseconds is written to the second
+ok 74 - a due date with milliseconds is written to the second
+  ---
+  duration_ms: 63.926825
+  type: 'test'
+  ...
+# Subtest: every write landed and the final read failed: the verb succeeds and records what it wrote
+ok 75 - every write landed and the final read failed: the verb succeeds and records what it wrote
+  ---
+  duration_ms: 37.114019
+  type: 'test'
+  ...
+# Subtest: some writes landed and the final read failed: write-uncertain, and nothing is recorded
+ok 76 - some writes landed and the final read failed: write-uncertain, and nothing is recorded
+  ---
+  duration_ms: 14.619195
+  type: 'test'
+  ...
+# Subtest: a create whose final read fails succeeds and records task.created
+ok 77 - a create whose final read fails succeeds and records task.created
+  ---
+  duration_ms: 32.420325
+  type: 'test'
+  ...
+# Subtest: an edit between the last write and the final read shows as external on the next poll
+ok 78 - an edit between the last write and the final read shows as external on the next poll
+  ---
+  duration_ms: 34.851743
+  type: 'test'
+  ...
+# Subtest: task.created is recorded when a later label write fails
+ok 79 - task.created is recorded when a later label write fails
+  ---
+  duration_ms: 10.803743
+  type: 'test'
+  ...
+# Subtest: every verb acts in an installed child, on the child's own board
+ok 80 - every verb acts in an installed child, on the child's own board
+  ---
+  duration_ms: 41.977682
+  type: 'test'
+  ...
+# Subtest: task.create refuses a project outside the tree, a watch-only one and one that is not an id
+ok 81 - task.create refuses a project outside the tree, a watch-only one and one that is not an id
+  ---
+  duration_ms: 22.713698
+  type: 'test'
+  ...
+# Subtest: with descendants off, task.create takes only the root project
+ok 82 - with descendants off, task.create takes only the root project
+  ---
+  duration_ms: 11.216167
+  type: 'test'
+  ...
+# Subtest: a verb on a watch-only task, or a relation to one, refuses with project-not-installed
+ok 83 - a verb on a watch-only task, or a relation to one, refuses with project-not-installed
+  ---
+  duration_ms: 15.686458
+  type: 'test'
+  ...
+# Subtest: a related task is read before authority: in another project than its ref names, task-moved; missing, task-not-found
+ok 84 - a related task is read before authority: in another project than its ref names, task-moved; missing, task-not-found
+  ---
+  duration_ms: 21.530086
+  type: 'test'
+  ...
+# Subtest: a task moved to another project of the tree refuses its old ref with task-moved and takes its new one
+ok 85 - a task moved to another project of the tree refuses its old ref with task-moved and takes its new one
+  ---
+  duration_ms: 14.767496
+  type: 'test'
+  ...
+# Subtest: a verb and a create in the child record the final read of the child's board
+ok 86 - a verb and a create in the child record the final read of the child's board
+  ---
+  duration_ms: 14.054079
+  type: 'test'
+  ...
+# Subtest: a create in the child whose final read fails records the child's todo bucket
+ok 87 - a create in the child whose final read fails records the child's todo bucket
+  ---
+  duration_ms: 33.155996
+  type: 'test'
+  ...
+# Subtest: a lost answer in the child is settled by a re-read of the child's board
+ok 88 - a lost answer in the child is settled by a re-read of the child's board
+  ---
+  duration_ms: 18.780476
+  type: 'test'
+  ...
+# Subtest: every related task is read before authority, not only the first
+ok 89 - every related task is read before authority, not only the first
+  ---
+  duration_ms: 14.326976
+  type: 'test'
+  ...
+1..89
+# tests 89
+# suites 0
+# pass 88
+# fail 1
+# cancelled 0
+# skipped 0
+# todo 0
+# duration_ms 980.81047
diff --git a/agents/researcher/work/queue-71/r1/out/mutants-stdout.txt b/agents/researcher/work/queue-71/r1/out/mutants-stdout.txt
new file mode 100644
index 000000000..cf42e4e23
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/mutants-stdout.txt
@@ -0,0 +1,5 @@
+r3-confirm-4xx killed rc=1 start=2026-10-13T14:48:04Z end=2026-10-13T14:48:05Z # tests 89 # pass 88 # fail 1
+    not ok: a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+r3-confirm-error-on-throw killed rc=1 start=2026-10-13T14:48:05Z end=2026-10-13T14:48:06Z # tests 89 # pass 88 # fail 1
+    not ok: a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+restored packages/tasks/src/sync.mjs sha256 35e79dc2ac56af8047bef11a5aaf5f1d37ce3e11513f9968454053068f59ebcf OK
diff --git a/agents/researcher/work/queue-71/r1/out/mutants-summary.txt b/agents/researcher/work/queue-71/r1/out/mutants-summary.txt
new file mode 100644
index 000000000..cf42e4e23
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/mutants-summary.txt
@@ -0,0 +1,5 @@
+r3-confirm-4xx killed rc=1 start=2026-10-13T14:48:04Z end=2026-10-13T14:48:05Z # tests 89 # pass 88 # fail 1
+    not ok: a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+r3-confirm-error-on-throw killed rc=1 start=2026-10-13T14:48:05Z end=2026-10-13T14:48:06Z # tests 89 # pass 88 # fail 1
+    not ok: a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+restored packages/tasks/src/sync.mjs sha256 35e79dc2ac56af8047bef11a5aaf5f1d37ce3e11513f9968454053068f59ebcf OK
diff --git a/agents/researcher/work/queue-71/r1/out/node-bus.log b/agents/researcher/work/queue-71/r1/out/node-bus.log
new file mode 100644
index 000000000..8153f5a21
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/node-bus.log
@@ -0,0 +1,700 @@
+TAP version 13
+# Subtest: agents: a held role with nothing on the bus is idle; a role session gets the claim rows unchanged
+ok 1 - agents: a held role with nothing on the bus is idle; a role session gets the claim rows unchanged
+  ---
+  duration_ms: 29.404392
+  type: 'test'
+  ...
+# Subtest: agents: working counts what the holder received and has not read; routed alone is not working
+ok 2 - agents: working counts what the holder received and has not read; routed alone is not working
+  ---
+  duration_ms: 24.388082
+  type: 'test'
+  ...
+# Subtest: agents: input needed is the role's open decisions routed to the human, gated first, then oldest
+ok 3 - agents: input needed is the role's open decisions routed to the human, gated first, then oldest
+  ---
+  duration_ms: 41.095966
+  type: 'test'
+  ...
+# Subtest: agents: working wins over waiting, and Input needed still shows
+ok 4 - agents: working wins over waiting, and Input needed still shows
+  ---
+  duration_ms: 21.103277
+  type: 'test'
+  ...
+# Subtest: agents: a new holder doesn't inherit what the last one left unread; decisions stay with the role
+ok 5 - agents: a new holder doesn't inherit what the last one left unread; decisions stay with the role
+  ---
+  duration_ms: 29.086999
+  type: 'test'
+  ...
+# Subtest: agents: state reads only its own business, even for a run name both use
+ok 6 - agents: state reads only its own business, even for a run name both use
+  ---
+  duration_ms: 21.29204
+  type: 'test'
+  ...
+# Subtest: launch identity is stamped, payload identity is refused and stale holder cannot send
+ok 7 - launch identity is stamped, payload identity is refused and stale holder cannot send
+  ---
+  duration_ms: 28.153702
+  type: 'test'
+  ...
+# Subtest: decision classes route from policy; gated resolution is human-only, choice and target must match
+ok 8 - decision classes route from policy; gated resolution is human-only, choice and target must match
+  ---
+  duration_ms: 29.101524
+  type: 'test'
+  ...
+# Subtest: claim exclusion, holder release, gated revoke and rerouting to a new holder are atomic
+ok 9 - claim exclusion, holder release, gated revoke and rerouting to a new holder are atomic
+  ---
+  duration_ms: 33.031822
+  type: 'test'
+  ...
+# Subtest: launch events require a human CLI capability; generic emit cannot forge authority events
+ok 10 - launch events require a human CLI capability; generic emit cannot forge authority events
+  ---
+  duration_ms: 19.817481
+  type: 'test'
+  ...
+# Subtest: within-role decisions close atomically and invalid options or blocking omissions refuse
+ok 11 - within-role decisions close atomically and invalid options or blocking omissions refuse
+  ---
+  duration_ms: 25.074911
+  type: 'test'
+  ...
+# Subtest: observer capabilities read human inbox but cannot mutate or forge launch identity
+ok 12 - observer capabilities read human inbox but cannot mutate or forge launch identity
+  ---
+  duration_ms: 20.872654
+  type: 'test'
+  ...
+# Subtest: task action subjects and linked decision trail are complete and ordered
+ok 13 - task action subjects and linked decision trail are complete and ordered
+  ---
+  duration_ms: 30.519334
+  type: 'test'
+  ...
+# Subtest: launch binding is durable and reconnecting requires the identical trusted record
+ok 14 - launch binding is durable and reconnecting requires the identical trusted record
+  ---
+  duration_ms: 13.571008
+  type: 'test'
+  ...
+# Subtest: business isolation includes inherited object names and cross-business message references
+ok 15 - business isolation includes inherited object names and cross-business message references
+  ---
+  duration_ms: 23.959227
+  type: 'test'
+  ...
+# Subtest: authority never transfers between action, run, target, unresolved or replaced role holder
+ok 16 - authority never transfers between action, run, target, unresolved or replaced role holder
+  ---
+  duration_ms: 27.468671
+  type: 'test'
+  ...
+# Subtest: task projection uses schema current view, skipping earlier and equal-start polls
+ok 17 - task projection uses schema current view, skipping earlier and equal-start polls
+  ---
+  duration_ms: 14.469312
+  type: 'test'
+  ...
+# Subtest: revocation permanently bars the old run from reclaiming first, including after broker restart
+ok 18 - revocation permanently bars the old run from reclaiming first, including after broker restart
+  ---
+  duration_ms: 19.536403
+  type: 'test'
+  ...
+# Subtest: empty message references refuse before storage; refusal-evidence failure stays a typed error
+ok 19 - empty message references refuse before storage; refusal-evidence failure stays a typed error
+  ---
+  duration_ms: 11.665092
+  type: 'test'
+  ...
+# Subtest: both arbiters require human resolution when their cross-role route is themselves
+ok 20 - both arbiters require human resolution when their cross-role route is themselves
+  ---
+  duration_ms: 20.53905
+  type: 'test'
+  ...
+# Subtest: the shipped coder role raises task.reassign as gated to the human; task.scope.change stays cross-role
+ok 21 - the shipped coder role raises task.reassign as gated to the human; task.scope.change stays cross-role
+  ---
+  duration_ms: 17.56469
+  type: 'test'
+  ...
+# Subtest: S1 adapter takes resolved limits and refs, rejects mismatched instance, never mutates input
+ok 22 - S1 adapter takes resolved limits and refs, rejects mismatched instance, never mutates input
+  ---
+  duration_ms: 2.752409
+  type: 'test'
+  ...
+# Subtest: credentials gives human and reader sessions service, account, role, date and state, and nothing else
+ok 23 - credentials gives human and reader sessions service, account, role, date and state, and nothing else
+  ---
+  duration_ms: 20.969979
+  type: 'test'
+  ...
+# Subtest: credentials refuses a role session and any argument
+ok 24 - credentials refuses a role session and any argument
+  ---
+  duration_ms: 19.490276
+  type: 'test'
+  ...
+# Subtest: credentials reads the status on each request, and a broker without one lists none
+ok 25 - credentials reads the status on each request, and a broker without one lists none
+  ---
+  duration_ms: 24.262478
+  type: 'test'
+  ...
+# Subtest: the broker process answers credentials without a token, path or variable in its output, log or store
+ok 26 - the broker process answers credentials without a token, path or variable in its output, log or store
+  ---
+  duration_ms: 80.104856
+  type: 'test'
+  ...
+# Subtest: only validated broker references load; returned data and exceptions cannot expose a known token
+ok 27 - only validated broker references load; returned data and exceptions cannot expose a known token
+  ---
+  duration_ms: 4.733079
+  type: 'test'
+  ...
+# Subtest: bad file modes, symlinks, repository/data paths, malformed tokens and missing dates refuse
+ok 28 - bad file modes, symlinks, repository/data paths, malformed tokens and missing dates refuse
+  ---
+  duration_ms: 4.008375
+  type: 'test'
+  ...
+# Subtest: expiry refuses use and env references never become client data
+ok 29 - expiry refuses use and env references never become client data
+  ---
+  duration_ms: 0.900712
+  type: 'test'
+  ...
+# Subtest: S1 parsed service refs work, service mismatch refuses, Gitea rotation due is a warning state
+ok 30 - S1 parsed service refs work, service mismatch refuses, Gitea rotation due is a warning state
+  ---
+  duration_ms: 2.062545
+  type: 'test'
+  ...
+# Subtest: opaque tokens shorter than 16 characters refuse before use
+ok 31 - opaque tokens shorter than 16 characters refuse before use
+  ---
+  duration_ms: 0.43369
+  type: 'test'
+  ...
+# Subtest: roles lists open decisions routed to a role, and never the human inbox
+ok 32 - roles lists open decisions routed to a role, and never the human inbox
+  ---
+  duration_ms: 37.367827
+  type: 'test'
+  ...
+# Subtest: recent lists closed decisions of every route, newest close first, with how each closed
+ok 33 - recent lists closed decisions of every route, newest close first, with how each closed
+  ---
+  duration_ms: 33.698975
+  type: 'test'
+  ...
+# Subtest: recent gives the closing event as closed when the decision was seen first
+ok 34 - recent gives the closing event as closed when the decision was seen first
+  ---
+  duration_ms: 36.302512
+  type: 'test'
+  ...
+# Subtest: recent stops at RECENT rows
+ok 35 - recent stops at RECENT rows
+  ---
+  duration_ms: 112.381059
+  type: 'test'
+  ...
+# Subtest: every row is what trail already gives the same session for that id
+ok 36 - every row is what trail already gives the same session for that id
+  ---
+  duration_ms: 39.417836
+  type: 'test'
+  ...
+# Subtest: a role session can't list decisions, so it can't read another role's open decisions
+ok 37 - a role session can't list decisions, so it can't read another role's open decisions
+  ---
+  duration_ms: 38.923681
+  type: 'test'
+  ...
+# Subtest: decisions is a read: one business, strict arguments, nothing written
+ok 38 - decisions is a read: one business, strict arguments, nothing written
+  ---
+  duration_ms: 34.929545
+  type: 'test'
+  ...
+# Subtest: endLaunch writes session.ended, releases the run claim and kills its capabilities
+ok 39 - endLaunch writes session.ended, releases the run claim and kills its capabilities
+  ---
+  duration_ms: 24.898626
+  type: 'test'
+  ...
+# Subtest: endLaunch refuses an unknown run, a second end and a rebind of the ended run
+ok 40 - endLaunch refuses an unknown run, a second end and a rebind of the ended run
+  ---
+  duration_ms: 14.111017
+  type: 'test'
+  ...
+# Subtest: a restarted broker refuses to rebind an ended run; a refused rebind leaves the run unbound
+ok 41 - a restarted broker refuses to rebind an ended run; a refused rebind leaves the run unbound
+  ---
+  duration_ms: 18.644999
+  type: 'test'
+  ...
+# Subtest: endLaunch leaves a claim another run took alone
+ok 42 - endLaunch leaves a claim another run took alone
+  ---
+  duration_ms: 20.834805
+  type: 'test'
+  ...
+# Subtest: refuse records action.refused against the caller with the code only
+ok 43 - refuse records action.refused against the caller with the code only
+  ---
+  duration_ms: 16.75486
+  type: 'test'
+  ...
+# Subtest: launches off refuses role.launch with launch-revoked until launches on
+ok 44 - launches off refuses role.launch with launch-revoked until launches on
+  ---
+  duration_ms: 21.337677
+  type: 'test'
+  ...
+# Subtest: broker process: launch ops authorize role.launch, record refusals and end runs
+ok 45 - broker process: launch ops authorize role.launch, record refusals and end runs
+  ---
+  duration_ms: 98.224744
+  type: 'test'
+  ...
+# Subtest: human proof binds CLI entry, process start and nonce; agents and incomplete ancestry refuse
+ok 46 - human proof binds CLI entry, process start and nonce; agents and incomplete ancestry refuse
+  ---
+  duration_ms: 3.785214
+  type: 'test'
+  ...
+# Subtest: process reader gets own kernel identity without exposing environment values
+ok 47 - process reader gets own kernel identity without exposing environment values
+  ---
+  duration_ms: 0.708465
+  type: 'test'
+  ...
+# Subtest: EACCES ancestor environments skip only markers; commands and registered launches still refuse
+ok 48 - EACCES ancestor environments skip only markers; commands and registered launches still refuse
+  ---
+  duration_ms: 0.931919
+  type: 'test'
+  ...
+# Subtest: real pid 1 remains inspectable when its environment is protected
+ok 49 - real pid 1 remains inspectable when its environment is protected
+  ---
+  duration_ms: 0.378096
+  type: 'test'
+  ...
+# Subtest: within-role sends cite an open gated launch decision without spending it or naming it in grants
+ok 50 - within-role sends cite an open gated launch decision without spending it or naming it in grants
+  ---
+  duration_ms: 31.467452
+  type: 'test'
+  ...
+# Subtest: missing and foreign-business citations refuse and roll back message and grant
+ok 51 - missing and foreign-business citations refuse and roll back message and grant
+  ---
+  duration_ms: 21.397751
+  type: 'test'
+  ...
+# Subtest: cross-role sends still need a matching resolved decision and consume it once
+ok 52 - cross-role sends still need a matching resolved decision and consume it once
+  ---
+  duration_ms: 31.69782
+  type: 'test'
+  ...
+# Subtest: messages pages all human messages newest first without writes; read state is history
+ok 53 - messages pages all human messages newest first without writes; read state is history
+  ---
+  duration_ms: 347.390182
+  type: 'test'
+  ...
+# Subtest: messages refuses role callers, invalid paging and cursors outside this human feed
+ok 54 - messages refuses role callers, invalid paging and cursors outside this human feed
+  ---
+  duration_ms: 18.565262
+  type: 'test'
+  ...
+# Subtest: views messages passes only its paging arguments through
+ok 55 - views messages passes only its paging arguments through
+  ---
+  duration_ms: 0.230416
+  type: 'test'
+  ...
+# Subtest: messages isolates businesses even with a valid foreign cursor
+ok 56 - messages isolates businesses even with a valid foreign cursor
+  ---
+  duration_ms: 10.260464
+  type: 'test'
+  ...
+# Subtest: broker process binds trusted launches, offers reader capabilities, refuses human mutation, closes cleanly
+ok 57 - broker process binds trusted launches, offers reader capabilities, refuses human mutation, closes cleanly
+  ---
+  duration_ms: 88.499191
+  type: 'test'
+  ...
+# Subtest: startup token refusal returns safe code without value or partial listening broker
+ok 58 - startup token refusal returns safe code without value or partial listening broker
+  ---
+  duration_ms: 55.806194
+  type: 'test'
+  ...
+# Subtest: loaded fixture token is absent from socket replies and SQLite, including refusal evidence
+ok 59 - loaded fixture token is absent from socket replies and SQLite, including refusal evidence
+  ---
+  duration_ms: 73.825352
+  type: 'test'
+  ...
+# Subtest: killed broker leaves an explicit stale lock; another process cannot silently reclaim it
+ok 60 - killed broker leaves an explicit stale lock; another process cannot silently reclaim it
+  ---
+  duration_ms: 93.548139
+  type: 'test'
+  ...
+# Subtest: trusted host registers later launches; socket clients never have a registration verb
+ok 61 - trusted host registers later launches; socket clients never have a registration verb
+  ---
+  duration_ms: 56.121313
+  type: 'test'
+  ...
+# Subtest: runtime excludes declared project roots even when host supplies no repoRoots
+ok 62 - runtime excludes declared project roots even when host supplies no repoRoots
+  ---
+  duration_ms: 37.978129
+  type: 'test'
+  ...
+# Subtest: a refused launch binding leaves the broker and existing capabilities alive; bad protocol stops it
+ok 63 - a refused launch binding leaves the broker and existing capabilities alive; bad protocol stops it
+  ---
+  duration_ms: 55.354582
+  type: 'test'
+  ...
+# Subtest: a reply is listed for human and reader sessions, and its delivery rows are unchanged
+ok 64 - a reply is listed for human and reader sessions, and its delivery rows are unchanged
+  ---
+  duration_ms: 43.544135
+  type: 'test'
+  ...
+# Subtest: the reply still reaches the human message.receive (mosaic talk) after the Console showed it
+ok 65 - the reply still reaches the human message.receive (mosaic talk) after the Console showed it
+  ---
+  duration_ms: 22.976264
+  type: 'test'
+  ...
+# Subtest: a message with no reply gives an empty list
+ok 66 - a message with no reply gives an empty list
+  ---
+  duration_ms: 19.763755
+  type: 'test'
+  ...
+# Subtest: replies come oldest first and stop at REPLIES
+ok 67 - replies come oldest first and stop at REPLIES
+  ---
+  duration_ms: 93.997141
+  type: 'test'
+  ...
+# Subtest: an id that names no message in the business refuses message-not-found, and the refusal is recorded
+ok 68 - an id that names no message in the business refuses message-not-found, and the refusal is recorded
+  ---
+  duration_ms: 33.267248
+  type: 'test'
+  ...
+# Subtest: strict arguments; a role session gets human-required
+ok 69 - strict arguments; a role session gets human-required
+  ---
+  duration_ms: 33.780563
+  type: 'test'
+  ...
+# Subtest: views passes replies through as one call with the id
+ok 70 - views passes replies through as one call with the id
+  ---
+  duration_ms: 0.239359
+  type: 'test'
+  ...
+# Subtest: requirements returns the committed list to a holding role session, and no one else
+ok 71 - requirements returns the committed list to a holding role session, and no one else
+  ---
+  duration_ms: 29.706814
+  type: 'test'
+  ...
+# Subtest: the PM's prd.approve with no task, when no id fits, reaches the human's Inbox
+ok 72 - the PM's prd.approve with no task, when no id fits, reaches the human's Inbox
+  ---
+  duration_ms: 18.864349
+  type: 'test'
+  ...
+# Subtest: the PM's non-blocking role.revoke with no task reaches the human's Inbox
+ok 73 - the PM's non-blocking role.revoke with no task reaches the human's Inbox
+  ---
+  duration_ms: 21.736755
+  type: 'test'
+  ...
+# Subtest: v3b prototype refusals, views and append-only mutations
+ok 74 - v3b prototype refusals, views and append-only mutations
+  ---
+  duration_ms: 94.540081
+  type: 'test'
+  ...
+# Subtest: gated approval authorizes once, survives store reopen, and fresh approval works
+ok 75 - gated approval authorizes once, survives store reopen, and fresh approval works
+  ---
+  duration_ms: 35.29223
+  type: 'test'
+  ...
+# Subtest: another run cannot consume an approval; a failed check leaves it usable
+ok 76 - another run cannot consume an approval; a failed check leaves it usable
+  ---
+  duration_ms: 27.487187
+  type: 'test'
+  ...
+# Subtest: two scheduled callers have exactly one grant and one consumed refusal
+ok 77 - two scheduled callers have exactly one grant and one consumed refusal
+  ---
+  duration_ms: 18.592453
+  type: 'test'
+  ...
+# Subtest: failed commit rolls consumption back; cross-role consumes and within-role stays reusable
+ok 78 - failed commit rolls consumption back; cross-role consumes and within-role stays reusable
+  ---
+  duration_ms: 42.955639
+  type: 'test'
+  ...
+# Subtest: class drift gated to cross-role refuses before consumption
+ok 79 - class drift gated to cross-role refuses before consumption
+  ---
+  duration_ms: 26.871433
+  type: 'test'
+  ...
+# Subtest: class drift cross-role to gated refuses before consumption
+ok 80 - class drift cross-role to gated refuses before consumption
+  ---
+  duration_ms: 27.161389
+  type: 'test'
+  ...
+# Subtest: class drift gated to within-role refuses before consumption
+ok 81 - class drift gated to within-role refuses before consumption
+  ---
+  duration_ms: 22.625194
+  type: 'test'
+  ...
+# Subtest: class drift cross-role to within-role refuses before consumption
+ok 82 - class drift cross-role to within-role refuses before consumption
+  ---
+  duration_ms: 25.887318
+  type: 'test'
+  ...
+# Subtest: class drift within-role to gated refuses before consumption
+ok 83 - class drift within-role to gated refuses before consumption
+  ---
+  duration_ms: 20.414737
+  type: 'test'
+  ...
+# Subtest: class drift within-role to cross-role refuses before consumption
+ok 84 - class drift within-role to cross-role refuses before consumption
+  ---
+  duration_ms: 19.083114
+  type: 'test'
+  ...
+# Subtest: message.send consumes approval and prevents a later send or authorize
+ok 85 - message.send consumes approval and prevents a later send or authorize
+  ---
+  duration_ms: 19.984908
+  type: 'test'
+  ...
+# Subtest: role.revoke consumes approval and prevents a later revoke or authorize
+ok 86 - role.revoke consumes approval and prevents a later revoke or authorize
+  ---
+  duration_ms: 20.262932
+  type: 'test'
+  ...
+# Subtest: creates private WAL store and excludes a second writer until explicit close
+ok 87 - creates private WAL store and excludes a second writer until explicit close
+  ---
+  duration_ms: 18.586235
+  type: 'test'
+  ...
+# Subtest: rollback is atomic and schema metadata is checked against trusted DDL, not just itself
+ok 88 - rollback is atomic and schema metadata is checked against trusted DDL, not just itself
+  ---
+  duration_ms: 19.823934
+  type: 'test'
+  ...
+# Subtest: existing empty database and symlink runtime directory refuse, never initialize over damage
+ok 89 - existing empty database and symlink runtime directory refuse, never initialize over damage
+  ---
+  duration_ms: 20.148158
+  type: 'test'
+  ...
+# Subtest: crash during a transaction recovers no partial event after explicit fixture-only lock removal
+ok 90 - crash during a transaction recovers no partial event after explicit fixture-only lock removal
+  ---
+  duration_ms: 62.573919
+  type: 'test'
+  ...
+# Subtest: writer refuses mixed at/read_at forms atomically, even through trusted SQL helpers
+ok 91 - writer refuses mixed at/read_at forms atomically, even through trusted SQL helpers
+  ---
+  duration_ms: 13.713291
+  type: 'test'
+  ...
+# Subtest: async transactions refuse before invoking their function
+ok 92 - async transactions refuse before invoking their function
+  ---
+  duration_ms: 10.245812
+  type: 'test'
+  ...
+# Subtest: the session default fits under the reply cap and pages to the end with next
+ok 93 - the session default fits under the reply cap and pages to the end with next
+  ---
+  duration_ms: 182.297793
+  type: 'test'
+  ...
+# Subtest: the short form: no description, assignees by role, every listed key
+ok 94 - the short form: no description, assignees by role, every listed key
+  ---
+  duration_ms: 80.842934
+  type: 'test'
+  ...
+# Subtest: each filter, alone and together, and total counts every match
+ok 95 - each filter, alone and together, and total counts every match
+  ---
+  duration_ms: 148.772203
+  type: 'test'
+  ...
+# Subtest: task_ref reads one task in full, as the whole board has it
+ok 96 - task_ref reads one task in full, as the whole board has it
+  ---
+  duration_ms: 86.059366
+  type: 'test'
+  ...
+# Subtest: no arguments is the read the Console and mosaic make, unchanged
+ok 97 - no arguments is the read the Console and mosaic make, unchanged
+  ---
+  duration_ms: 68.410868
+  type: 'test'
+  ...
+# Subtest: an unknown argument or a bad value refuses invalid-request
+ok 98 - an unknown argument or a bad value refuses invalid-request
+  ---
+  duration_ms: 287.824699
+  type: 'test'
+  ...
+# Subtest: tasks rows carry priority and requirement for every session that reads tasks
+ok 99 - tasks rows carry priority and requirement for every session that reads tasks
+  ---
+  duration_ms: 46.590704
+  type: 'test'
+  ...
+# Subtest: the read model rebuilds from the ledger: a reopened bus and each trail give the same values
+ok 100 - the read model rebuilds from the ledger: a reopened bus and each trail give the same values
+  ---
+  duration_ms: 51.526953
+  type: 'test'
+  ...
+# Subtest: an empty business and one with snapshots but no task.created read cleanly
+ok 101 - an empty business and one with snapshots but no task.created read cleanly
+  ---
+  duration_ms: 12.286622
+  type: 'test'
+  ...
+# Subtest: recordTask keeps sync reads and a role write apart
+ok 102 - recordTask keeps sync reads and a role write apart
+  ---
+  duration_ms: 24.793842
+  type: 'test'
+  ...
+# Subtest: read_at must be one canonical UTC format, so the projection compares strings safely
+ok 103 - read_at must be one canonical UTC format, so the projection compares strings safely
+  ---
+  duration_ms: 11.673334
+  type: 'test'
+  ...
+# Subtest: a bad entry refuses the whole record
+ok 104 - a bad entry refuses the whole record
+  ---
+  duration_ms: 11.412186
+  type: 'test'
+  ...
+# Subtest: taskView reads the projection for one business
+ok 105 - taskView reads the projection for one business
+  ---
+  duration_ms: 13.571013
+  type: 'test'
+  ...
+# Subtest: requestTask hands only a holder and a task verb to the handler, and records refusals
+ok 106 - requestTask hands only a holder and a task verb to the handler, and records refusals
+  ---
+  duration_ms: 23.931202
+  type: 'test'
+  ...
+# Subtest: the server sends task verbs to the adapter with its own timeout; other verbs stay synchronous
+ok 107 - the server sends task verbs to the adapter with its own timeout; other verbs stay synchronous
+  ---
+  duration_ms: 275.781364
+  type: 'test'
+  ...
+# Subtest: without an adapter the server refuses every task verb
+ok 108 - without an adapter the server refuses every task verb
+  ---
+  duration_ms: 15.732511
+  type: 'test'
+  ...
+# Subtest: the runtime refuses an invalid adapter and closes a valid one
+ok 109 - the runtime refuses an invalid adapter and closes a valid one
+  ---
+  duration_ms: 16.680061
+  type: 'test'
+  ...
+# Subtest: the process loads the S3 adapter from plain-data trackers
+ok 110 - the process loads the S3 adapter from plain-data trackers
+  ---
+  duration_ms: 98.450111
+  type: 'test'
+  ...
+# Subtest: socket capability stamps launch identity; shared views use wire, no SQL client
+ok 111 - socket capability stamps launch identity; shared views use wire, no SQL client
+  ---
+  duration_ms: 31.177984
+  type: 'test'
+  ...
+# Subtest: two wire claims serialize; a lost reply never automatically retries
+ok 112 - two wire claims serialize; a lost reply never automatically retries
+  ---
+  duration_ms: 23.843443
+  type: 'test'
+  ...
+# Subtest: malformed, oversized and identity-forging envelopes refuse without echoing input
+ok 113 - malformed, oversized and identity-forging envelopes refuse without echoing input
+  ---
+  duration_ms: 12.08672
+  type: 'test'
+  ...
+# Subtest: client preserves UTF-8 when a response divides a multibyte character
+ok 114 - client preserves UTF-8 when a response divides a multibyte character
+  ---
+  duration_ms: 12.689802
+  type: 'test'
+  ...
+# Subtest: committed mutation followed by dropped reply reports unknown and is never retried
+ok 115 - committed mutation followed by dropped reply reports unknown and is never retried
+  ---
+  duration_ms: 15.867775
+  type: 'test'
+  ...
+1..115
+# tests 115
+# suites 0
+# pass 115
+# fail 0
+# cancelled 0
+# skipped 0
+# todo 0
+# duration_ms 1009.541259
diff --git a/agents/researcher/work/queue-71/r1/out/node-tasks.log b/agents/researcher/work/queue-71/r1/out/node-tasks.log
new file mode 100644
index 000000000..30e809265
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/node-tasks.log
@@ -0,0 +1,544 @@
+TAP version 13
+# Subtest: the boot config is checked before anything starts
+ok 1 - the boot config is checked before anything starts
+  ---
+  duration_ms: 15.373002
+  type: 'test'
+  ...
+# Subtest: a business with no tracker entry refuses task verbs
+ok 2 - a business with no tracker entry refuses task verbs
+  ---
+  duration_ms: 16.54793
+  type: 'test'
+  ...
+# Subtest: credential.expiring and .expired are recorded once per instance
+ok 3 - credential.expiring and .expired are recorded once per instance
+  ---
+  duration_ms: 32.903372
+  type: 'test'
+  ...
+# Subtest: a token file that changes on disk records credential.changed
+ok 4 - a token file that changes on disk records credential.changed
+  ---
+  duration_ms: 13.718499
+  type: 'test'
+  ...
+# Subtest: autostart polls, reconciles and retries a startup the tracker was down for
+ok 5 - autostart polls, reconciles and retries a startup the tracker was down for
+  ---
+  duration_ms: 36.388706
+  type: 'test'
+  ...
+# Subtest: a refusal a restart must clear is not retried by the poll
+ok 6 - a refusal a restart must clear is not retried by the poll
+  ---
+  duration_ms: 11.731165
+  type: 'test'
+  ...
+# Subtest: a poll that fires while two are queued is dropped
+ok 7 - a poll that fires while two are queued is dropped
+  ---
+  duration_ms: 14.792592
+  type: 'test'
+  ...
+# Subtest: close waits for a running verb and refuses one that has not started
+ok 8 - close waits for a running verb and refuses one that has not started
+  ---
+  duration_ms: 43.160938
+  type: 'test'
+  ...
+# Subtest: the bundled Vikunja is the pinned upstream image the runbook names
+ok 9 - the bundled Vikunja is the pinned upstream image the runbook names
+  ---
+  duration_ms: 1.122644
+  type: 'test'
+  ...
+# Subtest: every published port is on 127.0.0.1, and no secret is in the file
+ok 10 - every published port is on 127.0.0.1, and no secret is in the file
+  ---
+  duration_ms: 0.422977
+  type: 'test'
+  ...
+# Subtest: the committed table loads frozen; its verbs are the broker task verbs
+ok 11 - the committed table loads frozen; its verbs are the broker task verbs
+  ---
+  duration_ms: 1.401375
+  type: 'test'
+  ...
+# Subtest: the schema refuses any other shape
+ok 12 - the schema refuses any other shape
+  ---
+  duration_ms: 3.119555
+  type: 'test'
+  ...
+# Subtest: each verb refuses, asks for a decision and writes as the table and the role files say
+ok 13 - each verb refuses, asks for a decision and writes as the table and the role files say
+  ---
+  duration_ms: 161.876618
+  type: 'test'
+  ...
+# Subtest: the list holds the PRD's requirement ids, in its order, at its version
+ok 14 - the list holds the PRD's requirement ids, in its order, at its version
+  ---
+  duration_ms: 5.015825
+  type: 'test'
+  ...
+# Subtest: the committed list loads frozen
+ok 15 - the committed list loads frozen
+  ---
+  duration_ms: 0.503693
+  type: 'test'
+  ...
+# Subtest: the schema refuses any other shape
+ok 16 - the schema refuses any other shape
+  ---
+  duration_ms: 2.487269
+  type: 'test'
+  ...
+# Subtest: the fake answers each route with the statuses and shapes Vikunja v2.7.0 sent
+ok 17 - the fake answers each route with the statuses and shapes Vikunja v2.7.0 sent
+  ---
+  duration_ms: 98.430176
+  type: 'test'
+  ...
+# Subtest: the project list holds recorded project objects in the recorded list envelope
+ok 18 - the project list holds recorded project objects in the recorded list envelope
+  ---
+  duration_ms: 10.372561
+  type: 'test'
+  ...
+# Subtest: the recorded task bodies pass the checks S3 applies to every read
+ok 19 - the recorded task bodies pass the checks S3 applies to every read
+  ---
+  duration_ms: 0.447974
+  type: 'test'
+  ...
+# Subtest: the client works against the fake over real HTTP with the platform fetch
+ok 20 - the client works against the fake over real HTTP with the platform fetch
+  ---
+  duration_ms: 100.522814
+  type: 'test'
+  ...
+# Subtest: a correct install starts, and the first reconcile records tasks that already exist
+ok 21 - a correct install starts, and the first reconcile records tasks that already exist
+  ---
+  duration_ms: 25.186592
+  type: 'test'
+  ...
+# Subtest: verbs refuse while a business is starting and after startup refused it
+ok 22 - verbs refuse while a business is starting and after startup refused it
+  ---
+  duration_ms: 16.584152
+  type: 'test'
+  ...
+# Subtest: startup refuses a token that can do more than its role needs
+ok 23 - startup refuses a token that can do more than its role needs
+  ---
+  duration_ms: 43.957423
+  type: 'test'
+  ...
+# Subtest: startup refuses an unsupported version and flags an untested one
+ok 24 - startup refuses an unsupported version and flags an untested one
+  ---
+  duration_ms: 30.417189
+  type: 'test'
+  ...
+# Subtest: startup refuses a board that the runbook did not install
+ok 25 - startup refuses a board that the runbook did not install
+  ---
+  duration_ms: 39.769969
+  type: 'test'
+  ...
+# Subtest: startup refuses a project the sync bot cannot read
+ok 26 - startup refuses a project the sync bot cannot read
+  ---
+  duration_ms: 10.619813
+  type: 'test'
+  ...
+# Subtest: startup refuses a configured label the pm bot cannot see
+ok 27 - startup refuses a configured label the pm bot cannot see
+  ---
+  duration_ms: 9.55423
+  type: 'test'
+  ...
+# Subtest: startup refuses an expired credential and a missing sync credential
+ok 28 - startup refuses an expired credential and a missing sync credential
+  ---
+  duration_ms: 15.33053
+  type: 'test'
+  ...
+# Subtest: an unreachable tracker refuses with tracker-unavailable
+ok 29 - an unreachable tracker refuses with tracker-unavailable
+  ---
+  duration_ms: 6.047777
+  type: 'test'
+  ...
+# Subtest: an edit in the UI is recorded once, with the fields that changed
+ok 30 - an edit in the UI is recorded once, with the fields that changed
+  ---
+  duration_ms: 81.651407
+  type: 'test'
+  ...
+# Subtest: a move between open buckets is seen on the board, though updated does not change
+ok 31 - a move between open buckets is seen on the board, though updated does not change
+  ---
+  duration_ms: 40.65521
+  type: 'test'
+  ...
+# Subtest: a person's comment is counted and a bot's is not
+ok 32 - a person's comment is counted and a bot's is not
+  ---
+  duration_ms: 72.160882
+  type: 'test'
+  ...
+# Subtest: the hourly reconcile catches a comment through comment_count
+ok 33 - the hourly reconcile catches a comment through comment_count
+  ---
+  duration_ms: 75.571973
+  type: 'test'
+  ...
+# Subtest: a task closed in the UI leaves the open view with its done bucket
+ok 34 - a task closed in the UI leaves the open view with its done bucket
+  ---
+  duration_ms: 114.050267
+  type: 'test'
+  ...
+# Subtest: a task that leaves the board is recorded as deleted, moved or out of reach
+ok 35 - a task that leaves the board is recorded as deleted, moved or out of reach
+  ---
+  duration_ms: 62.861146
+  type: 'test'
+  ...
+# Subtest: a poll that read before a verb wrote does not overwrite the verb
+ok 36 - a poll that read before a verb wrote does not overwrite the verb
+  ---
+  duration_ms: 33.650755
+  type: 'test'
+  ...
+# Subtest: a tracker fault during a tick is reported and the next tick catches up
+ok 37 - a tracker fault during a tick is reported and the next tick catches up
+  ---
+  duration_ms: 36.726265
+  type: 'test'
+  ...
+# Subtest: a malformed answer refuses the tick with tracker-shape
+ok 38 - a malformed answer refuses the tick with tracker-shape
+  ---
+  duration_ms: 13.16272
+  type: 'test'
+  ...
+# Subtest: no token value reaches the database, the log or a refusal
+ok 39 - no token value reaches the database, the log or a refusal
+  ---
+  duration_ms: 45.451924
+  type: 'test'
+  ...
+# Subtest: the first look at a task counts only comments inside the window
+ok 40 - the first look at a task counts only comments inside the window
+  ---
+  duration_ms: 34.839318
+  type: 'test'
+  ...
+# Subtest: with descendants off the adapter reads no project list and no child, as in v0
+ok 41 - with descendants off the adapter reads no project list and no child, as in v0
+  ---
+  duration_ms: 48.293931
+  type: 'test'
+  ...
+# Subtest: a three-level tree is read per project, and a project outside it is not
+ok 42 - a three-level tree is read per project, and a project outside it is not
+  ---
+  duration_ms: 22.741197
+  type: 'test'
+  ...
+# Subtest: a watch-only child keeps the business ready, and a failed child read refuses as transient
+ok 43 - a watch-only child keeps the business ready, and a failed child read refuses as transient
+  ---
+  duration_ms: 25.584269
+  type: 'test'
+  ...
+# Subtest: a root without the installed buckets refuses with descendants on
+ok 44 - a root without the installed buckets refuses with descendants on
+  ---
+  duration_ms: 10.829984
+  type: 'test'
+  ...
+# Subtest: the project list refuses without read_all, and skips archived projects and saved filters
+ok 45 - the project list refuses without read_all, and skips archived projects and saved filters
+  ---
+  duration_ms: 25.614391
+  type: 'test'
+  ...
+# Subtest: descendants must be true or false
+ok 46 - descendants must be true or false
+  ---
+  duration_ms: 10.090398
+  type: 'test'
+  ...
+# Subtest: a move inside the tree is a moved tombstone plus a snapshot under the new ref
+ok 47 - a move inside the tree is a moved tombstone plus a snapshot under the new ref
+  ---
+  duration_ms: 63.257475
+  type: 'test'
+  ...
+# Subtest: a child made after startup joins the tree at the next reconcile
+ok 48 - a child made after startup joins the tree at the next reconcile
+  ---
+  duration_ms: 53.464339
+  type: 'test'
+  ...
+# Subtest: a child that leaves the tree has its tasks recorded missing, and the ledger keeps their history
+ok 49 - a child that leaves the tree has its tasks recorded missing, and the ledger keeps their history
+  ---
+  duration_ms: 59.379728
+  type: 'test'
+  ...
+# Subtest: a child deleted or hidden between reconciles drops out at the next tick, and the root still refuses
+ok 50 - a child deleted or hidden between reconciles drops out at the next tick, and the root still refuses
+  ---
+  duration_ms: 80.722644
+  type: 'test'
+  ...
+# Subtest: a reconcile that fails keeps the previous tree
+ok 51 - a reconcile that fails keeps the previous tree
+  ---
+  duration_ms: 30.274693
+  type: 'test'
+  ...
+# Subtest: a closed task the cursor missed takes its own project's done bucket
+ok 52 - a closed task the cursor missed takes its own project's done bucket
+  ---
+  duration_ms: 33.556879
+  type: 'test'
+  ...
+# Subtest: closed tasks read at startup take their own project's done bucket
+ok 53 - closed tasks read at startup take their own project's done bucket
+  ---
+  duration_ms: 11.304826
+  type: 'test'
+  ...
+# Subtest: a closed task the cursor returns takes its own project's done bucket
+ok 54 - a closed task the cursor returns takes its own project's done bucket
+  ---
+  duration_ms: 34.47553
+  type: 'test'
+  ...
+# Subtest: a task moved out of a child that leaves the tree at the same time is moved, not no-access
+ok 55 - a task moved out of a child that leaves the tree at the same time is moved, not no-access
+  ---
+  duration_ms: 33.855457
+  type: 'test'
+  ...
+# Subtest: a task in a bucket outside the five keeps its raw bucket id
+ok 56 - a task in a bucket outside the five keeps its raw bucket id
+  ---
+  duration_ms: 29.916939
+  type: 'test'
+  ...
+# Subtest: bucket 0 on a watch-only task never becomes a bucket or a state
+ok 57 - bucket 0 on a watch-only task never becomes a bucket or a state
+  ---
+  duration_ms: 78.180511
+  type: 'test'
+  ...
+# Subtest: a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+ok 58 - a child whose kanban view is deleted still refuses the tick, and the reconcile reads it watch-only
+  ---
+  duration_ms: 55.83525
+  type: 'test'
+  ...
+# Subtest: a child dropped at a tick comes back at the next reconcile once the bot reads it again
+ok 59 - a child dropped at a tick comes back at the next reconcile once the bot reads it again
+  ---
+  duration_ms: 35.039514
+  type: 'test'
+  ...
+# Subtest: task.create needs a recorded human request and a requirement id
+ok 60 - task.create needs a recorded human request and a requirement id
+  ---
+  duration_ms: 40.038164
+  type: 'test'
+  ...
+# Subtest: task.create takes only an approved requirement id
+ok 61 - task.create takes only an approved requirement id
+  ---
+  duration_ms: 26.188455
+  type: 'test'
+  ...
+# Subtest: task.create cites the request a received message carries, not the message's id
+ok 62 - task.create cites the request a received message carries, not the message's id
+  ---
+  duration_ms: 20.204034
+  type: 'test'
+  ...
+# Subtest: only labels named in the business file can be written
+ok 63 - only labels named in the business file can be written
+  ---
+  duration_ms: 28.889903
+  type: 'test'
+  ...
+# Subtest: task.schedule sets and clears a due date and relations
+ok 64 - task.schedule sets and clears a due date and relations
+  ---
+  duration_ms: 35.511892
+  type: 'test'
+  ...
+# Subtest: assign and reassign move the role bots and record task.assigned
+ok 65 - assign and reassign move the role bots and record task.assigned
+  ---
+  duration_ms: 52.985464
+  type: 'test'
+  ...
+# Subtest: task.update.assigned is for the assignee and records task.state
+ok 66 - task.update.assigned is for the assignee and records task.state
+  ---
+  duration_ms: 45.462286
+  type: 'test'
+  ...
+# Subtest: a wrong expected digest records task.conflict and writes nothing
+ok 67 - a wrong expected digest records task.conflict and writes nothing
+  ---
+  duration_ms: 18.246971
+  type: 'test'
+  ...
+# Subtest: a cross-role verb needs a resolved decision, used once
+ok 68 - a cross-role verb needs a resolved decision, used once
+  ---
+  duration_ms: 20.824764
+  type: 'test'
+  ...
+# Subtest: task.close needs a verdict; after it every verb refuses with task-done
+ok 69 - task.close needs a verdict; after it every verb refuses with task-done
+  ---
+  duration_ms: 18.669837
+  type: 'test'
+  ...
+# Subtest: a lost answer is settled by a re-read and never retried
+ok 70 - a lost answer is settled by a re-read and never retried
+  ---
+  duration_ms: 22.962907
+  type: 'test'
+  ...
+# Subtest: a create whose answer is lost is reported uncertain, and the poll finds the task
+ok 71 - a create whose answer is lost is reported uncertain, and the poll finds the task
+  ---
+  duration_ms: 34.833368
+  type: 'test'
+  ...
+# Subtest: a task the sync bot cannot read refuses and records nothing
+ok 72 - a task the sync bot cannot read refuses and records nothing
+  ---
+  duration_ms: 14.189432
+  type: 'test'
+  ...
+# Subtest: verbs and polls for one business run one at a time
+ok 73 - verbs and polls for one business run one at a time
+  ---
+  duration_ms: 87.217517
+  type: 'test'
+  ...
+# Subtest: a due date with milliseconds is written to the second
+ok 74 - a due date with milliseconds is written to the second
+  ---
+  duration_ms: 63.3084
+  type: 'test'
+  ...
+# Subtest: every write landed and the final read failed: the verb succeeds and records what it wrote
+ok 75 - every write landed and the final read failed: the verb succeeds and records what it wrote
+  ---
+  duration_ms: 36.767011
+  type: 'test'
+  ...
+# Subtest: some writes landed and the final read failed: write-uncertain, and nothing is recorded
+ok 76 - some writes landed and the final read failed: write-uncertain, and nothing is recorded
+  ---
+  duration_ms: 14.892048
+  type: 'test'
+  ...
+# Subtest: a create whose final read fails succeeds and records task.created
+ok 77 - a create whose final read fails succeeds and records task.created
+  ---
+  duration_ms: 33.229455
+  type: 'test'
+  ...
+# Subtest: an edit between the last write and the final read shows as external on the next poll
+ok 78 - an edit between the last write and the final read shows as external on the next poll
+  ---
+  duration_ms: 34.927009
+  type: 'test'
+  ...
+# Subtest: task.created is recorded when a later label write fails
+ok 79 - task.created is recorded when a later label write fails
+  ---
+  duration_ms: 12.685295
+  type: 'test'
+  ...
+# Subtest: every verb acts in an installed child, on the child's own board
+ok 80 - every verb acts in an installed child, on the child's own board
+  ---
+  duration_ms: 40.250043
+  type: 'test'
+  ...
+# Subtest: task.create refuses a project outside the tree, a watch-only one and one that is not an id
+ok 81 - task.create refuses a project outside the tree, a watch-only one and one that is not an id
+  ---
+  duration_ms: 23.879872
+  type: 'test'
+  ...
+# Subtest: with descendants off, task.create takes only the root project
+ok 82 - with descendants off, task.create takes only the root project
+  ---
+  duration_ms: 11.825594
+  type: 'test'
+  ...
+# Subtest: a verb on a watch-only task, or a relation to one, refuses with project-not-installed
+ok 83 - a verb on a watch-only task, or a relation to one, refuses with project-not-installed
+  ---
+  duration_ms: 17.946104
+  type: 'test'
+  ...
+# Subtest: a related task is read before authority: in another project than its ref names, task-moved; missing, task-not-found
+ok 84 - a related task is read before authority: in another project than its ref names, task-moved; missing, task-not-found
+  ---
+  duration_ms: 23.395865
+  type: 'test'
+  ...
+# Subtest: a task moved to another project of the tree refuses its old ref with task-moved and takes its new one
+ok 85 - a task moved to another project of the tree refuses its old ref with task-moved and takes its new one
+  ---
+  duration_ms: 14.305279
+  type: 'test'
+  ...
+# Subtest: a verb and a create in the child record the final read of the child's board
+ok 86 - a verb and a create in the child record the final read of the child's board
+  ---
+  duration_ms: 12.940352
+  type: 'test'
+  ...
+# Subtest: a create in the child whose final read fails records the child's todo bucket
+ok 87 - a create in the child whose final read fails records the child's todo bucket
+  ---
+  duration_ms: 33.588967
+  type: 'test'
+  ...
+# Subtest: a lost answer in the child is settled by a re-read of the child's board
+ok 88 - a lost answer in the child is settled by a re-read of the child's board
+  ---
+  duration_ms: 21.070278
+  type: 'test'
+  ...
+# Subtest: every related task is read before authority, not only the first
+ok 89 - every related task is read before authority, not only the first
+  ---
+  duration_ms: 15.963255
+  type: 'test'
+  ...
+1..89
+# tests 89
+# suites 0
+# pass 89
+# fail 0
+# cancelled 0
+# skipped 0
+# todo 0
+# duration_ms 983.468677
diff --git a/agents/researcher/work/queue-71/r1/out/test-auth.log b/agents/researcher/work/queue-71/r1/out/test-auth.log
new file mode 100644
index 000000000..ac2287606
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-auth.log
@@ -0,0 +1,17 @@
+OK    status with missing harness credential exits 3 and still lists accounts
+OK    status reports harness credential (read-only) + mosaic accounts
+OK    api key material never reaches output
+OK    oauth token material never reaches output
+OK    unparseable credential file exits 2
+OK    symlinked credential file exits 4
+OK    env-side credential names reported
+OK    env var values never reach output
+OK    accounts without an accounts dir reports none and creates nothing
+OK    accounts lists files and marks the active one
+OK    loose account perms flagged in listing
+OK    agent --auth with missing account file refuses (exit 4)
+OK    agent --auth with non-0600 account file refuses
+OK    agent --auth with invalid account name refuses
+OK    auth.sh without valid config refuses
+
+selftest: 15 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/out/test-conductor.log b/agents/researcher/work/queue-71/r1/out/test-conductor.log
new file mode 100644
index 000000000..62afc513c
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-conductor.log
@@ -0,0 +1,23 @@
+On branch refactor
+Your branch is up to date with 'origin/refactor'.
+
+nothing to commit, working tree clean
+OK    dry-run: allowed change, exit 0, nothing committed (exit 0)
+OK    dry-run committed nothing
+OK    apply: allowed change exits 0 (exit 0)
+OK    apply: attribution in commit subject
+OK    apply: target tree clean after commit
+OK    disallowed path refused (exit 1)
+OK    disallowed path: target untouched
+OK    syntax gate refused broken .mjs (exit 1)
+OK    syntax gate: target untouched
+OK    suite failure refused (exit 1)
+OK    suite failure: target reverted to clean
+OK    disabled policy refused (exit 2)
+OK    disabled policy: target untouched
+OK    failed run refused (exit 1)
+OK    failed run: target untouched
+OK    missing run exits 4 (exit 4)
+OK    invalid policy exits 2 (exit 2)
+
+selftest: 17 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/out/test-config.log b/agents/researcher/work/queue-71/r1/out/test-config.log
new file mode 100644
index 000000000..76c05ef74
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-config.log
@@ -0,0 +1,26 @@
+OK    absent adapter defaults to pi
+OK    adapter mock validates (exit 0)
+OK    unsupported adapter exits 2 (exit 2)
+OK    env exports adapter
+OK    bootstrap creates default when absent (exit 0)
+OK    bootstrap wrote config file
+OK    bootstrap is idempotent on existing config (exit 0)
+OK    bootstrap did not rewrite existing config
+OK    validate missing config exits 3 (exit 3)
+OK    malformed JSON exits 2 (exit 2)
+OK    unsupported configVersion exits 2 (exit 2)
+OK    unknown top-level key exits 2 (exit 2)
+OK    unknown execution key exits 2 (exit 2)
+OK    unsupported backend exits 2 (exit 2)
+OK    unsupported environment exits 2 (exit 2)
+OK    relative dataRoot exits 2 (exit 2)
+OK    non-canonical dataRoot exits 2 (exit 2)
+OK    filesystem root dataRoot exits 2 (exit 2)
+OK    home directory dataRoot exits 2 (exit 2)
+OK    dataRoot containing config dir exits 2 (exit 2)
+OK    control character in provider exits 2 (exit 2)
+OK    symlinked config file exits 2 (exit 2)
+OK    env exports resolve correctly
+OK    failed validation modified nothing
+
+selftest: 24 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/out/test-discord.log b/agents/researcher/work/queue-71/r1/out/test-discord.log
new file mode 100644
index 000000000..b013f9ffa
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-discord.log
@@ -0,0 +1,70 @@
+toolchain: node v26.8.1
+
+OK    syntax: packages/discord/src/approvals.mjs
+OK    syntax: packages/discord/src/authorize.mjs
+OK    syntax: packages/discord/src/binding.mjs
+OK    syntax: packages/discord/src/cli.mjs
+OK    syntax: packages/discord/src/connector.mjs
+OK    syntax: packages/discord/src/context.mjs
+OK    syntax: packages/discord/src/engine-pi.mjs
+OK    syntax: packages/discord/src/errors.mjs
+OK    syntax: packages/discord/src/gateway.mjs
+OK    syntax: packages/discord/src/git.mjs
+OK    syntax: packages/discord/src/journal.mjs
+OK    syntax: packages/discord/src/notify.mjs
+OK    syntax: packages/discord/src/rest.mjs
+OK    syntax: packages/discord/src/setspark.mjs
+OK    syntax: packages/discord/src/tools.mjs
+OK    syntax: packages/discord/src/web.mjs
+OK    syntax: packages/discord/bin/git-credential.mjs
+OK    syntax: packages/discord/extension/tools.mjs
+OK    syntax: packages/discord/tests/approvals.test.mjs
+OK    syntax: packages/discord/tests/authorize.test.mjs
+OK    syntax: packages/discord/tests/binding.test.mjs
+OK    syntax: packages/discord/tests/connector.test.mjs
+OK    syntax: packages/discord/tests/context.test.mjs
+OK    syntax: packages/discord/tests/engine.test.mjs
+OK    syntax: packages/discord/tests/fake-pi.mjs
+OK    syntax: packages/discord/tests/gateway.test.mjs
+OK    syntax: packages/discord/tests/git.test.mjs
+OK    syntax: packages/discord/tests/helpers.mjs
+OK    syntax: packages/discord/tests/journal.test.mjs
+OK    syntax: packages/discord/tests/notify.test.mjs
+OK    syntax: packages/discord/tests/recover.test.mjs
+OK    syntax: packages/discord/tests/rest.test.mjs
+OK    syntax: packages/discord/tests/setspark.test.mjs
+OK    syntax: packages/discord/tests/tools.test.mjs
+OK    syntax: packages/discord/tests/web.test.mjs
+OK    syntax: packages/discord/fixtures/claim-worker.mjs
+OK    syntax: packages/discord/fixtures/legacy-owner-worker.mjs
+OK    syntax: scripts/discord.sh
+OK    syntax: scripts/discord-service.sh
+OK    packages/discord declares no dependencies
+OK    no bot-token-shaped string in packages/discord
+OK    fixture binding uses placeholder ids only
+OK    fixture binding validates
+OK    real pi with the extension exposes exactly list_dir, read_file, search and no built-in tool
+OK    real pi with a writable root exposes exactly the three reads plus write_file and edit_file, and writes nothing at start
+OK    real pi with a web key exposes the three reads plus web_fetch and web_search, and no write tool without a writable root
+OK    real pi with a git root exposes the reads, writes and the four git verbs, commits nothing at start, and never shows the token
+OK    real pi with protocol vault adds reserve_id to the git verbs
+OK    real pi with a setspark key exposes the reads and the eight record verbs, no counters, and never shows the key
+OK    real pi refuses a git key on a read-only root (fail closed)
+OK    real pi with the pilot flags (--no-tools) exposes no tool at all
+OK    real pi exits non-zero without MOSAIC_DISCORD_TOOLS: no session, no tools (fail closed)
+OK    a failing nested test fails the run under a parent runner's NODE_TEST_CONTEXT
+OK    node --test packages/discord/tests/ (ℹ pass 178)
+OK    scripts/discord.sh --help exits 0
+OK    scripts/discord.sh check without a binding exits 4
+OK    scripts/discord.sh recover without a binding exits 4
+OK    scripts/discord.sh reload without a binding exits 4
+OK    scripts/discord-service.sh without a command exits 4
+OK    service unit renders with the repository path, a supervised run as the main process, exit 3 never retried, and reload as SIGHUP
+OK    service install writes the rendered unit (0644) and leaves no temp file
+OK    service install a second time reports unchanged
+OK    systemd-analyze verify accepts the rendered unit
+OK    service uninstall removes the unit file
+OK    service install with an unknown flag exits 4
+OK    service install with USER unset finishes and names the account for lingering
+
+discord suite: 66 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/out/test-extension-package.log b/agents/researcher/work/queue-71/r1/out/test-extension-package.log
new file mode 100644
index 000000000..dad5eeaf0
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-extension-package.log
@@ -0,0 +1,21 @@
+OK    initial ordinary-file install
+OK    installed tree matches canonical source
+OK    installed tree has no symlinks
+OK    check detects installation drift
+OK    sync refuses to overwrite installation drift
+OK    check detects an extra destination file
+OK    check detects an extra destination directory
+OK    check rejects a destination symlink
+OK    sync accepts a canonical source update
+OK    updated installation matches canonical source
+scripts/test-extension-package.sh: line 14: 955374 Killed                     "$@" > /dev/null 2>&1
+OK    forced interruption kills the replacing process
+OK    next invocation recovers old consistent installation
+OK    interrupted replacement rolled back
+OK    sync succeeds after interruption recovery
+OK    unlocked stale lock file does not block
+OK    active lock refuses a concurrent sync
+OK    source symlink fails closed
+OK    nested second entrypoint fails closed
+
+extension package selftest: 18 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/out/test-foundation.log b/agents/researcher/work/queue-71/r1/out/test-foundation.log
new file mode 100644
index 000000000..6338e1e6d
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-foundation.log
@@ -0,0 +1,53 @@
+toolchain: node v26.8.1, python 3.12.8, jsonschema 4.26.0
+
+OK    syntax: scripts/foundation-inspect.mjs
+OK    syntax: scripts/foundation/strict-json.mjs
+OK    syntax: scripts/foundation/canonical.mjs
+OK    syntax: scripts/foundation/resolve.mjs
+OK    syntax: scripts/foundation/validate-record.mjs
+OK    syntax: scripts/foundation/fixtures/build-fixtures.mjs
+OK    syntax: scripts/foundation/canonical.test.mjs
+OK    syntax: scripts/foundation/cli.test.mjs
+OK    syntax: scripts/foundation/fixtures.test.mjs
+OK    syntax: scripts/foundation/resolve.test.mjs
+OK    syntax: scripts/foundation/strict-json.test.mjs
+OK    syntax: scripts/foundation/verify-schema.py (ast only; no bytecode written)
+OK    fixture generator runs
+OK    checked-in fixtures/bundles equal a fresh generation
+OK    checked-in fixtures/raw equal a fresh generation
+OK    checked-in fixtures/index.json equal a fresh generation
+OK    checked-in demo bundles equal a fresh generation
+OK    a failing nested test fails the run under a parent runner's NODE_TEST_CONTEXT
+OK    node --test scripts/foundation/ (ℹ pass 80)
+OK    differential schema oracle: PASS: differential schema oracle (finite corpus; compatibility evidence, not equivalence proof)
+      platform witness: strftime('%Y') for year 999 -> '999' (pinned checker refuses years 0001..0999)
+      node v26.8.1; corpus 1568 records (38 pinned fixtures, 478 unique bundle records, 1052 typeCase/mutation/lexical cases)
+      schema column: agree-valid 540, agree-invalid 991, DISAGREEMENTS 0; strict-only (parser-bound) cases: 27; unsupported-kind records not schema-assessed by the inspector: 10
+      profile column (schema-valid records only): profile-valid 510, profile-invalid 30
+      profile refusals asserted: 30 schema-agreed-valid records refused only by the strict typed-string profile (rule profile-pattern-mismatch), 12 declared by name; 73 named probes verified against declared schema/profile columns
+OK    oracle: zero schema-column disagreements with the pinned checker
+OK    oracle: strict-only profile refusals are counted and asserted
+OK    demo: permitted read preview exits 0 (exit 0)
+OK    demo: permitted file.change preview exits 0 (exit 0)
+OK    demo: assignment.change proposal is unresolved (exit 3) (exit 3)
+OK    demo: revoked registration is refused (exit 3) (exit 3)
+OK    demo: message is not authority (exit 3) (exit 3)
+OK    usage: no arguments exits 2 (exit 2)
+OK    io: missing file exits 4 (exit 4)
+OK    io: directory exits 4 (exit 4)
+OK    io: symlink exits 4 (O_NOFOLLOW) (exit 4)
+OK    bound: oversize fixture exits 2 (exit 2)
+OK    profile: one final LF in a typed selection id is refused before admission (exit 2) (exit 2)
+OK    profile: two final LFs fail the schema pattern itself (exit 2) (exit 2)
+OK    profile: escaped newlines in free-form text stay allowed (exit 0) (exit 0)
+OK    profile refusal is invalid-request/profile-pattern-mismatch with selection and operation withheld, value not echoed
+OK    text output starts with the disclaimer
+OK    json output is valid JSON with result allowed and exactly the charter §7 fields
+OK    json golden matches byte-for-byte
+OK    sandboxed bundle run (env -i, PATH=/nonexistent) produced the unresolved proposal
+OK    sandbox inventory (path/type/size/mode/uid/gid/inode/mtime/sha256) unchanged by runs
+OK    canary never printed (bundle run and credential-file run)
+OK    a non-bundle JSON file is refused at the shape gate, not read into output
+OK    no field of the non-bundle file is echoed
+
+selftest: 44 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/out/test-queue.log b/agents/researcher/work/queue-71/r1/out/test-queue.log
new file mode 100644
index 000000000..d980d1b3d
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-queue.log
@@ -0,0 +1,35 @@
+toolchain: node v26.8.1, git version 2.55.0
+
+OK    syntax: packages/queue/src/cli.mjs
+OK    syntax: packages/queue/src/errors.mjs
+OK    syntax: packages/queue/src/io.mjs
+OK    syntax: packages/queue/src/lock.mjs
+OK    syntax: packages/queue/src/queue.mjs
+OK    syntax: packages/queue/src/review.mjs
+OK    syntax: packages/queue/src/store.mjs
+OK    syntax: packages/queue/tests/commit.test.mjs
+OK    syntax: packages/queue/tests/data.test.mjs
+OK    syntax: packages/queue/tests/dispatch.test.mjs
+OK    syntax: packages/queue/tests/helpers.mjs
+OK    syntax: packages/queue/tests/lock.test.mjs
+OK    syntax: packages/queue/tests/migration.test.mjs
+OK    syntax: packages/queue/tests/review.test.mjs
+OK    syntax: packages/queue/tests/store.test.mjs
+OK    syntax: packages/queue/tests/write.test.mjs
+OK    syntax: packages/queue/tests/fixtures/fake-gitea.mjs
+OK    syntax: packages/queue/tests/fixtures/kill-at.mjs
+OK    syntax: packages/queue/tests/fixtures/lock-child.mjs
+OK    syntax: packages/queue/tests/fixtures/mosaic-pre-a2.sh
+OK    syntax: scripts/queue-commit.sh
+OK    syntax: scripts/git-hooks/pre-commit
+OK    syntax: scripts/mosaic
+OK    queue-commit.sh, the guard and scripts/mosaic are executable
+OK    packages/queue declares no dependencies
+ℹ tests 150
+ℹ pass 150
+ℹ fail 0
+OK    node --test packages/queue/tests/
+OK    scripts/mosaic queue help
+skip  queue verify and render --check: this checkout (/mnt/storage/src/mosaic-stack/.claude/worktrees/agent-a44468a48c485520e) is not the queue's canonical root (/mnt/storage/src/mosaic-stack)
+
+queue suite: 27 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/out/test-release.log b/agents/researcher/work/queue-71/r1/out/test-release.log
new file mode 100644
index 000000000..7202b85d4
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-release.log
@@ -0,0 +1,16 @@
+OK    valid RELEASE resolves (exit 0)
+OK    invalid RELEASE exits 1 (exit 1)
+OK    missing RELEASE exits 1 (exit 1)
+OK    valid RELEASE leaves image tag consistent with version
+OK    status safe on empty state (exit 0)
+OK    status created no pointer
+OK    fault-injected activation refuses (exit 1)
+OK    refused activation wrote no pointer
+OK    refusal logged exactly once with valid fields
+OK    healthy activation succeeds (exit 0)
+OK    pointer written with valid fields
+OK    repeat activation succeeds (log grows) (exit 0)
+OK    log is append-only across activations
+OK    rollback without previous refuses (exit 1)
+
+selftest: 14 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/out/test-task.log b/agents/researcher/work/queue-71/r1/out/test-task.log
new file mode 100644
index 000000000..cd89fe9b2
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/out/test-task.log
@@ -0,0 +1,32 @@
+OK    valid task validates (exit 0)
+OK    unknown task key exits 2 (exit 2)
+OK    unsupported taskVersion exits 2 (exit 2)
+OK    invalid task id exits 2 (exit 2)
+OK    empty prompt exits 2 (exit 2)
+OK    NUL in expectExact exits 2 (exit 2)
+OK    out-of-range timeout exits 2 (exit 2)
+OK    missing mission file exits 4 (exit 4)
+OK    task with valid mission validates (exit 0)
+OK    invalid mission exits 2 (exit 2)
+OK    validate missing task exits 4 (exit 4)
+OK    validation does not modify the task file
+OK    prune dry-run exits 0 (exit 0)
+OK    dry-run deleted nothing
+OK    prune --keep=2 --yes removes oldest (exit 0)
+OK    kept exactly 2 newest runs
+OK    newest run kept, oldest pruned
+OK    append-only receipt written (3 entries)
+OK    sessions/workspaces untouched by prune
+OK    prune with invalid keep exits 4 (exit 4)
+skip adapter seam cases (docker daemon unavailable)
+skip workspace/capability cases (docker daemon unavailable)
+skip live task cases (docker unavailable)
+OK    onboard without name exits 4 (non-interactive) (exit 4)
+OK    onboard --name renders profile (exit 0)
+OK    profile written
+OK    canon structure: required filled, optional placeholdered
+OK    canon sections present
+skip live recall pair: user recall run, recalled user name (docker daemon unavailable)
+OK    no agent identity on headless run
+
+selftest: 26 passed, 0 failed
diff --git a/agents/researcher/work/queue-71/r1/run_group.py b/agents/researcher/work/queue-71/r1/run_group.py
new file mode 100644
index 000000000..a2a9f53f2
--- /dev/null
+++ b/agents/researcher/work/queue-71/r1/run_group.py
@@ -0,0 +1,55 @@
+#!/usr/bin/env python3
+"""Run one command in its own session and process group, with a timeout.
+
+usage: run_group.py TIMEOUT_S LOGFILE CWD -- CMD [ARGS...]
+
+Child stdout and stderr go to LOGFILE. On timeout the whole process group gets
+SIGTERM, then SIGKILL after 5 s. After the child exits, anything still left in
+its group is killed too, so nothing from one step outlives it.
+Exit status: the child's, or 124 on timeout.
+"""
+import os
+import signal
+import subprocess
+import sys
+
+
+def group_kill(pgid, sig):
+    try:
+        os.killpg(pgid, sig)
+    except ProcessLookupError:
+        pass
+
+
+def main():
+    timeout = float(sys.argv[1])
+    log = sys.argv[2]
+    cwd = sys.argv[3]
+    cmd = sys.argv[sys.argv.index('--') + 1:]
+    with open(log, 'w') as out:
+        p = subprocess.Popen(
+            cmd,
+            cwd=cwd,
+            stdout=out,
+            stderr=subprocess.STDOUT,
+            stdin=subprocess.DEVNULL,
+            start_new_session=True,
+        )
+        try:
+            rc = p.wait(timeout=timeout)
+        except subprocess.TimeoutExpired:
+            group_kill(p.pid, signal.SIGTERM)
+            try:
+                p.wait(timeout=5)
+            except subprocess.TimeoutExpired:
+                pass
+            group_kill(p.pid, signal.SIGKILL)
+            p.wait()
+            out.write(f"\nTIMEOUT after {timeout:.0f}s: process group {p.pid} killed\n")
+            rc = 124
+        group_kill(p.pid, signal.SIGKILL)
+    sys.exit(rc)
+
+
+if __name__ == '__main__':
+    main()