[noahgift]

Manual batch 2 (operator ruling, checkin-ruling-1615 item 1): three ready branches cherry-picked onto origin/main in one PR. Test-only or small refactors; no gate or workflow file touched. Each fold has pace + triage seats at its sha.

Folded:

• test: cover print_coverage_report in aprender-contracts-cli/src/commands/coverage.rs #5621 test(contracts-cli): pv coverage report rendering (+93)
• test: cover collect_partials_cdp in aprender-test-lib/src/validators.rs #5625 test(test-lib): collect_partials_cdp via partials_from_raw (+81/-23)
• test: cover generate_fkr_tests in aprender-ptx-debug/src/output/fkr_generator.rs #5624 test(ptx-debug): generate_fkr_tests output (+120/-3)

Ejected: #5761 (changelog_fragments.sh). The first make gate failed check_guards_are_wired (unwired guards 3 -> 4, NEW: changelog_fragments.sh). Reverting it makes that guard pass (232 scanned, 3 unwired). Wiring it needs a workflow edit, so it goes solo.

Gate status at open: the failing guard re-run green; a full make gate rerun is in flight (log /mnt/nvme-raid0/targets/batch2-gate2.log). Not armed until it is green and the seats are repeated at the PR head. Derived-file regen produced no diff.

keep-open: none
ont-delta: none

🤖 Generated with Claude Code

[noahgift]

seat 1, aprender-pace @6eeed7dcc589de56d46c6bb1409fa09dce49c27d

Repeat seat. Live head confirmed; 0 aprender-pace trailers. Net diff vs merge-base is 9 files: the three branches I seated, with their added/removed lines byte-identical to the branch diffs (#5621 @5b47a693d4, #5625 @2343f18977, #5624 @d91fb9e995). The #5761 content (scripts/changelog_fragments.sh, changelog.d/, .gitignore lines) is gone from the net diff; the eject commit 6eeed7d removes it. No workflow, gate or ci/sections.yml file touched; the only ci/ file is the new explicit-test-commands .cmd for the #5621 test.

Not verified: tests/CI not run or read.

[noahgift]

seat, aprender-triage-worker @6eeed7dcc589de56d46c6bb1409fa09dce49c27d APPROVE (repeat of my batch-2 seats at the PR head)

Own-trailer check: 0 of 6 commits carry a triage-worker trailer.
All 9 files in this PR are byte-identical (blob hash compared) to the versions I read and approved on the branches: 5b47a69 (#5621: .cmd fragment, pv contract YAML, coverage_report_render.rs), 2343f18 (#5625: lib.rs, validators.rs, validators_partials_5625_tests.rs), d91fb9e (#5624: ptx-debug contract YAML, bugs/mod.rs, fkr_generator.rs). The #5761 commit and the ejecting commit net to nothing in the file set (no changelog.d, no script, no .gitignore change).
Flag for the gate, not a code defect: two contract YAMLs gain a falsification line (FALSIFY-BIN-PV-011, FALSIFY-BIN-APRENDER-PTX-DEBUG-007) but contracts.nt, the witness and the README count are not in the file set; the contracts gate judges whether it needs a regen.
Not run: no local build or test; PR is BLOCKED at read, CI not judged. +294/-26 is over the 200-line step-5 cap (3 tickets); that is the cop's call.

[noahgift]

seat, aprender-triage-worker @9a5d783d2557b0104528ac2d609c3b6c271112ee APPROVE (re-seat; prior seat @6eeed7dcc5)

Own-trailer check: 0 commits carry a triage-worker trailer.
Delta in the PR's file set from my last seat: one file, scripts/tree_reader_tests.txt, +1 line aprender-contracts-cli --test coverage_report_render. It registers the new #5621 integration test target in the tree-reader list, in the sorted spot (after cli_integration, before ev11_lint_ratchets), and matches the test file's name. The other 9 files are unchanged from what I read. Not run; CI not judged.

[noahgift]

seat 1, aprender-pace @9a5d783d2557b0104528ac2d609c3b6c271112ee

Re-seat. 0 aprender-pace trailers. Delta vs my seat @6eeed7dcc5 = one commit, one line: scripts/tree_reader_tests.txt gains aprender-contracts-cli --test coverage_report_render, sorted into place. That file says "derived, never edit", so I ran scripts/check_tree_reader_tests.sh on this head in a throwaway worktree: PASS, "221 tree-reader target(s), registry equals the derived set". The line is the derived one, not a hand edit. Everything else is unchanged from my last seat.

Not verified: tests/CI not run or read.

[noahgift]

seat, aprender-triage-worker @d15e08d75249908e2e849f48bec95d94bddf53e2 APPROVE (re-seat; prior seat @9a5d783d25)

Own-trailer check: 0 commits carry a triage-worker trailer.
Delta in the PR's file set from my last seat: one line, in contracts/bin-aprender-contracts-cli--pv-v1.yaml, id FALSIFY-BIN-PV-011 -> FALSIFY-BIN-PV-006. Checked: origin/main has PV-001..005, this head has PV-001..006 each once, so the id is contiguous and unique; rule/prediction/test/mutation text unchanged. The other files are unchanged from what I read; the rest of the head delta is the main merge. Not run (a3 reports validate_contracts green locally; I did not repeat it); PR BLOCKED at read, CI not judged.

[noahgift]

seat 1, aprender-pace @d15e08d75249908e2e849f48bec95d94bddf53e2

Re-seat. 0 aprender-pace trailers. Delta vs my seat @9a5d783d25 = one fix commit (d179e13) plus one merge of origin/main. The PR's own diff (generated files excluded) differs in exactly one line: the new contract row id FALSIFY-BIN-PV-011 is now FALSIFY-BIN-PV-006 in contracts/bin-aprender-contracts-cli--pv-v1.yaml. The ids in that file are now PV-001..006 with no gap, which is what the data-integrity count (445 vs ceiling 444) needed. No BIN-PV-011 reference is left anywhere in the tree at this head, and contracts.nt carries neither id, so no regen was owed.

Not verified: validate_contracts / tests / CI not run or read; I did not re-run the 445 vs 444 count.

[github-actions]

§13.11 rung 1 — quorum shadow verdict

S13-SHADOW pr=5766 head=d15e08d75249908e2e849f48bec95d94bddf53e2 verdict=REFUSE class=Q1 arm_rc=1

Shadow mode: this records a verdict and merges nothing. A refusal
to arm is not a block (§13 adds zero rows to §7) — the pull request is
exactly as green as it was.