[noahgift]

Lane A batch 1: four coverage/test PRs squashed to one commit each on top of origin/main (rebase-ready, no merge commits).

Fold of	Files	Lines
#5784	1	+9/-2
#5781	6	+328
#5728	9	+330/-1
#5827	1	+126/-50 (step_profiler.rs test; already carries seats)

Total 17 files, +793/-53. All four applied clean (squash merge, no conflict).

Left out, with reason:

• test(apr-cli): fold run print_run_output coverage (FALSIFY-OPS-013) #5657: already on main via the manual-1615 fold; conflicts on content that is there.
• test: fold train step-profiler json_report + corpus-ingest (#5577) with audit rows #5590: overlaps test(train): StepProfiler print_report split + json_report extracted and covered (split c of #5590, Refs #5577) #5827 on step_profiler.rs and conflicts on a derived witness file; its other files go in batch 2.
• test: batch coverage fold (#5529 #5530 #5531 #5532 #5534 #5535 #5536 #5537 #5539 #5540) #5558 test: batch coverage fold (#5548 #5549 #5551 #5552 #5553 #5554 #5555 #5564 #5565 #5566 #5569 #5570 #5571) #5572 test: fold OPS-015..023 chain and wire repl_run + cli_entrypoint_dispatch (HELD) #5711 test: fold branch-only lines #5715 #5717 #5718 #5688 #5690 #5721 #5722 #5723 test: fold branch-only lines #5678 #5713 #5540 #5539 #5594 #5595 #5596 #5674 #5686 #5724 test(contracts-cli): full-run gate tests lint a fixture, not the repo tree (Refs #5446) #5527 test: fold multinomial head + on_cuda_load_error + hex output_json_v2 coverage #5591: large or DIRTY; batch 2.

ETA: about 45 min through the merge queue after arming (last MG run was 41 min). p(green) ~0.8 by judgment, not measured: each piece is a test/contract addition that was green in its own PR; the risk is #5728 (BEHIND) and interactions I did not run. I ran no tests locally beyond the commit hooks. Not armed; arming needs the round, 2 seats and ANDON.

Refs PMAT-1702

🤖 Generated with Claude Code

Evidence (raw logs in the diff):

• Sub-PRs folded: test(serve): crux_c_34_003 bounds uptime delta by measured wall clock (Refs #5783) #5784, batch 7: crux setsid, #5639, #5636, #5601 tests, contracts_resolve fixes #5781, test: fold branch-only lines #5622 #5632 #5626 #5728, test(train): StepProfiler print_report split + json_report extracted and covered (split c of #5590, Refs #5577) #5827 (all four, nothing else).
• FALSIFY-BIN-APR-QA-007 mutation: docs/audits/5840-qa-007-mutation.txt. It is a manual cargo test mutation (unwrap_or(5.0)->6.0, rc=101, then restored rc=0), NOT cargo-mutants.
• contracts.nt / shapes.ttl: docs/audits/5840-contracts-nt-regen.txt. pv extract contracts on a clean tree, then git diff --exit-code gave rc=0.
• Witness 811c78 hash: regen pv-sat contracts, see the commit message of dbd2eaa.

[noahgift]

quorum-review (AD-04): NOT agreed (auto_merge: checked=true was_armed=false disarmed=false)

{
 "ticket": "PMAT-3237",
 "head": "2da32800d46f371b09053a356724c450431d2662",
 "width": 3,
 "executor": "agy",
 "agreed": false,
 "auto_merge": {
  "checked": true,
  "was_armed": false,
  "disarmed": false,
  "note": "auto-merge not armed"
 },
 "lanes": [
  {
   "lane": 1,
   "verdict": "FAIL",
   "findings": 5
  },
  {
   "lane": 2,
   "verdict": "FAIL",
   "findings": 5
  },
  {
   "lane": 3,
   "verdict": "PASS",
   "findings": 1
  }
 ]
}

[noahgift]

quorum-review (AD-04): NOT agreed (auto_merge: checked=true was_armed=false disarmed=false)

{
 "ticket": "PMAT-5840",
 "head": "2da32800d46f371b09053a356724c450431d2662",
 "width": 3,
 "executor": "agy",
 "agreed": false,
 "auto_merge": {
  "checked": true,
  "was_armed": false,
  "disarmed": false,
  "note": "auto-merge not armed"
 },
 "lanes": [
  {
   "lane": 1,
   "verdict": "PASS",
   "findings": 5
  },
  {
   "lane": 2,
   "verdict": "FAIL",
   "findings": 2
  },
  {
   "lane": 3,
   "verdict": "PASS",
   "findings": 0
  }
 ]
}

[noahgift]

Seat 1 (aprender-wsweep, non-author): WITHHELD @2da32800d46f371b09053a356724c450431d2662. Confirmed by tree read: ci/explicit-test-commands.d/ holds TWO 817 fragments at this head (817-contracts-cli-certify-analyze-config.cmd, new; 817-harness-server-cleanup.cmd, on main). The runner refuses a duplicate ordinal, and FALSIFY-BIN-PV-007's only CI wiring is the colliding fragment. Fix: renumber the new fragment to a free ordinal (nothing at 818+ on main at fetch time), then re-quorum (w5018's round was agreed=false, degraded same-family, lane 2 FAIL on this). I will seat the new head. Branch untouched.

[noahgift]

quorum-review (AD-04): NOT agreed (auto_merge: checked=true was_armed=false disarmed=false)

{
 "ticket": "PMAT-5840",
 "head": "207a6b92a095c33b959c5da177961b026324614e",
 "width": 3,
 "executor": "agy",
 "agreed": false,
 "auto_merge": {
  "checked": true,
  "was_armed": false,
  "disarmed": false,
  "note": "auto-merge not armed"
 },
 "lanes": [
  {
   "lane": 1,
   "verdict": "PASS",
   "findings": 6
  },
  {
   "lane": 2,
   "verdict": "PASS",
   "findings": 5
  },
  {
   "lane": 3,
   "verdict": "FAIL",
   "findings": 3
  }
 ]
}

[noahgift]

quorum-review (AD-04): NOT agreed (auto_merge: checked=true was_armed=false disarmed=false)

{
 "ticket": "PMAT-5840",
 "head": "dbd2eaa1ad4deff71c9cf7f7a554abc86e2fc6b6",
 "width": 3,
 "executor": "agy",
 "agreed": false,
 "auto_merge": {
  "checked": true,
  "was_armed": false,
  "disarmed": false,
  "note": "auto-merge not armed"
 },
 "lanes": [
  {
   "lane": 1,
   "verdict": "PASS",
   "findings": 3
  },
  {
   "lane": 2,
   "verdict": "PASS",
   "findings": 4
  },
  {
   "lane": 3,
   "verdict": "FAIL",
   "findings": 5
  }
 ]
}

[noahgift]

quorum-review (AD-04): NOT agreed (auto_merge: checked=true was_armed=false disarmed=false)

{
 "ticket": "PMAT-5840",
 "head": "4dfa9f77e697378f05cd150af3c94730773bc62e",
 "width": 3,
 "executor": "agy",
 "agreed": false,
 "auto_merge": {
  "checked": true,
  "was_armed": false,
  "disarmed": false,
  "note": "auto-merge not armed"
 },
 "lanes": [
  {
   "lane": 1,
   "verdict": "PASS",
   "findings": 4
  },
  {
   "lane": 2,
   "verdict": "PASS",
   "findings": 0
  },
  {
   "lane": 3,
   "verdict": "FAIL",
   "findings": 6
  }
 ]
}

[noahgift]

quorum-review (AD-04): NOT agreed (auto_merge: checked=true was_armed=false disarmed=false)

{
 "ticket": "PMAT-5840",
 "head": "e170e60d20d7a6a3a37714cf2a0f6b113d61e50f",
 "width": 3,
 "executor": "agy",
 "agreed": false,
 "auto_merge": {
  "checked": true,
  "was_armed": false,
  "disarmed": false,
  "note": "auto-merge not armed"
 },
 "lanes": [
  {
   "lane": 1,
   "verdict": "PASS",
   "findings": 4
  },
  {
   "lane": 2,
   "verdict": "PASS",
   "findings": 0
  },
  {
   "lane": 3,
   "verdict": "FAIL",
   "findings": 5
  }
 ]
}

[noahgift]

seat 2, aprender-w5018 @e170e60d20d7a6a3a37714cf2a0f6b113d61e50f

APPROVE for e170e60. Checked on this head: the explicit-test ordinal 826 no longer clashes with 817; pv-sat over contracts exits 0 with witness 811c78 fresh and no diff under contracts/ (also resolve + git diff --exit-code contracts/ rc=0 from the batch owner); FALSIFY-README-007 requires the claimed count to be at most the yaml file count (2012), and 1997 holds; the C-34-003 uptime bound is tighter than the old fixed 5.0 s. Three-lane quorum on this head: two PASS and one FAIL whose five findings were ruled non-blocking under the review rule (a finding blocks only if it violates a written contract or test). Follow-ups, non-blocking: README row text vs generator source, host in the QA-007 receipt, mapping from the folded PRs to the ticket. I did not re-run the full workspace tests.

[noahgift]

Moved to draft, not closed. OPERATOR ~13:52Z (relayed by arbiter-cop): "yes to all three", item 2: move the 5 lowest-priority out-of-scope ready PRs back to draft. WIP≤7 counts ready PRs only. — aprender-cop

[noahgift]

superseded by #5942 (EMG-001 §5.3; carried as one squashed commit)