REPORT — phase6-scan-anchor-review
STEP: phase6-scan-anchor (Phase 6: the scan matches what the plans drive — F27)
REV: cc2ac82   FINAL: fa75ff06c2011745ebf9807dcd6c527e389d9574
CHANGED: docs/research/phase6-scan-anchor-review.md — this report. No review code rides: no finding needed a reproducer (every attack re-drove committed members, the deleted REV-riding probes on a REV scratch, or one-line scratch mutants diff-verified against `git show <rev>:…` before use), so no new test file is offered.
VERIFIED: REVIEW ACCEPTANCE run first (the method's only gated check): `scripts/check-scope cc2ac822d78f243f4c0dece9d3dff7e8f7ed5108 fa75ff06c2011745ebf9807dcd6c527e389d9574 docs/research/phase6-scan-anchor-brief.md` — exit 0, `claimed: tests/test_batch3_execution.py tests/test_review_phase6_probe_scope.py`, `tripped: none`, `scope: ok (3 path(s) in cc2ac82…fa75ff06…)`. The tree under review is clean (`nothing to commit, working tree clean` at HEAD `6f38af9`, whose diff to FINAL touches only the two report/brief docs files — the test and source trees are byte-identical between FINAL and the review tree, `git diff fa75ff0 6f38af9 --stat` names only `docs/research/phase6-scan-anchor-implementer.md` and `docs/research/phase6-scan-anchor-review-brief.md`). `git diff cc2ac82… fa75ff06… --stat` names exactly three paths — the two IN SCOPE test files plus `docs/research/phase6-scan-anchor-brief.md` (added by the planner's `84772ac` inside the range; the work commit itself touches only the two IN SCOPE files, `98 insertions(+), 135 deletions(-)`, including the deleted probe module) — no committed test edit outside IN SCOPE. The review brief's embedded copy (lines 20–177) is byte-identical to the committed `docs/research/phase6-scan-anchor-brief.md` (158 lines each) — no ORIGIN: BRIEF divergence rides. The work commit `fa75ff06` and the report commit `df21eaa` both carry `Actor: IMPLEMENTER`; this session wrote none of the commits under review; FINAL is the work commit, never the report commit.
VERIFIED: Q1(a) — the anchor is one source, the delivery's one question. At FINAL the constant `_PROBE_ARGV = ("/usr/bin/python3", "/work/probe.py")` sits at `tests/test_batch3_execution.py:36`; `_probe_plan` (`:61`), the F1 escape pin (`:248`) and the F20 racing pin (`:323`) all take `command=_PROBE_ARGV`; a grep over the delivered file finds no remaining plan-command tuple literal (the three REV literals at `:49`/`:231`/`:306` are gone), and a function-body diff of both pins REV→FINAL shows exactly the `command=…` substitution — no assertion change, no mark change. The pins keep shape and walls. Re-drive of the drift class against the delivered module, C1 (a second argv element in all three plans, `_PROBE_ARGV` untouched) on a diff-verified FINAL scratch (`/tmp/kilo/scan-anchor-final`, `git archive fa75ff0 | tar -x` with the shared `.venv` symlinked): `.venv/bin/python -m pytest tests/test_batch3_execution.py::test_every_probe_plan_command_still_names_the_scanned_pair tests/test_batch3_execution.py::test_the_scoped_scan_counts_a_live_pair_holder_and_not_an_argv_carrier -v` → `2 failed in 0.32s` — which member reddens on which assertion: the anchor member reddens on `assert command == _PROBE_ARGV` at `:358` (`plan command ('/usr/bin/python3', '-u', '/work/probe.py') drifted from the scanned pair …`, `At index 1 diff: '-u' != '/work/probe.py'` — the probe's measured class), the conjunction member reddens on `assert holder.pid in found` at `:395` (`the scoped scan missed a live workload-surface holder: []`, `assert 3775578 in []`). C2 (interpreter path moved to `/usr/bin/python3.13` in all three plans) on the same scratch: the anchor member reddens on the same assertion with `At index 0 diff: '/usr/bin/python3.13' != '/usr/bin/python3'`. Green at FINAL without drift: the batch3 battery `.venv/bin/python -m pytest tests/test_batch3_execution.py --durations=5` — `15 passed in 4.04s` (both absorbed members in; slowest five: `1.12s` racing pin, `0.91s` cancellation pin, `0.60s` deadline pin, `0.59s` F1 escape pin, `0.25s` limits pin). Attacked-and-not-falsified: the scanned pair is the driven pair, and the drift class is red at the delivered anchor.
VERIFIED: Q1(b) — the conjunction member states the selectivity in one scan at FINAL. The member (`tests/test_batch3_execution.py:363`, host-only guard `:368` — no `/sys/fs/cgroup/cgroup.controllers` visible → skip, the R08-type fresh root) makes one `_probe_pids()` call (`:389`, the delivered module's own scanner) and states all three halves on that single `found`: `assert carrier.pid not in found` (`:392`, the unrelated argv-carrier not counted), `assert holder.pid in found` (`:395`, the live workload-surface holder counted), `assert found == [holder.pid]` (`:398`, exactly the holder). Re-drive at the delivered module, no drift: `1 passed in 0.14s`; the same member re-driven red on the Q1(a) C1 drift above names `:395` as its failing line. The F24 shape's "not counted … while … still is" is one measure in one scan. Attacked-and-not-falsified.
VERIFIED: Q1(c) — the absorbed members' reds at REV, on the REV scratch (`/tmp/kilo/scan-anchor-rev`, `git archive cc2ac82 | tar -x`; `tests/test_batch3_execution.py`, `tests/test_review_phase6_probe_scope.py` and `src/orchestrator/sandbox.py` each diff-verified identical to `git show cc2ac82:…` before use; the shared `.venv` symlinked in). C1 drift (all three plan tuples → `("/usr/bin/python3", "-u", "/work/probe.py")`, `_PROBE_ARGV` `:184` untouched): `.venv/bin/python -m pytest tests/test_batch3_execution.py -q` → `13 passed in 3.81s` (every committed pin green over the blind scan, the sealed needle unchanged), and the riding probes `.venv/bin/python -m pytest tests/test_review_phase6_probe_scope.py -v` → `2 failed` — the anchor member on `assert command == _PROBE_ARGV` with `At index 1 diff: '-u' != '/work/probe.py'` and the conjunction member on `assert holder.pid in found` (`assert 3776401 in []`) — both the report's lines, replayed. The stacked battery: the same C1 scratch plus the weakened-capture mutant (one inserted `return []` at the head of `SandboxRunner._descendants`; `diff git show cc2ac82:src/orchestrator/sandbox.py` shows exactly one inserted line) — `.venv/bin/python -m pytest tests/test_batch3_execution.py::test_escaped_process_is_killed_without_a_cgroup tests/test_batch3_execution.py::test_racing_forker_and_double_forked_sleeper_die_at_teardown -v` → `2 passed in 1.68s` while `distinct drifted survivors alive: 123` (the escaped workload plus 122 racing-forker sleepers, argv `['/usr/bin/python3', '-u', '/work/probe.py']`), then `remaining after reap: 0` and `pair-matching set: []` throughout — the scan never sees the drifted shape, the pins' red class and their `finally` cleanups both stay blind, exactly the report's line. Contrast, the mutant alone with the C1 drift reverted: the same two nodes → `2 failed in 1.77s`, `assert survivors == []` (`assert [3776945] == []` on F1, a live racing-forker list on F20) — the mutant alone reddens both pins, so the stack's `2 passed` is the drift hiding the capture failure, not the mutant being green. Every guarantee attacked-and-not-falsified; no falsified class.
VERIFIED: Q2 — F29's absorb-both reading holds. The brief's absorption paragraph in isolation ("the seam-anchor member reddens on nothing — it asserts what the module now holds structurally"; the parenthetical "(both members absorbed or superseded; …)") does admit both readings if read alone. Reading (b) — the seam-anchor member superseded, only the conjunction member absorbed — does not stand against the rest of the brief: (i) the paragraph's own deletion branch takes its evidence from the red-evidence paragraph ("deletion needs no red (the absorbed members' reds below are the evidence)") and that paragraph names the anchor member's own red (`the anchor member's own red is the C1/C2 drift against the delivered anchor`) among the absorbed members' reds — under (b) the anchor member would not be an absorbed member and its red could not serve; (ii) the anchor paragraph unambiguously names the delivered form "the review probe's anchor member moved, not re-derived"; (iii) the GOAL's blanket "a plan-shape change either moves the scan with it or fails a committed member" fails under (b): the conjunction member's holder is built from `_probe_plan`'s plan only (verified in the delivered member — `_probe_plan(...)` feeds the holder), so a plan-shape change re-introducing a drifted literal at the F1/F20 pins' sites alone (`:248`/`:323`) would fail no committed member there — exactly the half-armed F27 hazard the finding says the anchor member exists to close; (iv) F27's inherited FIXED directs "a later brief absorbs these members into the pinned module" — both. The delivered module holds both members with the anchor member's assertion class unchanged: the probe's cross-module imports (`_PROBE_ARGV`, `_probe_pids`, `_probe_plan`) are gone because the member now lives in the module it scans; `assert command == _PROBE_ARGV` is unchanged in class (every plan command equals the scanned pair); the Name-node adaptation is the only semantic move — `ast.literal_eval(ast.Name(…))` raises `ValueError: malformed node or string` (reproduced here), so the verbatim move errors on the constant the plans now hold, and the delivered `globals().get(node.id) if isinstance(node, ast.Name) else ast.literal_eval(node)` resolves it; conjunction member moved verbatim (body identical to the probe's, host-only guard kept, no slow mark; the module-side `.venv`-free `_wait_cmdline` helper absorbed alongside). The module-side adjustments are exactly the two named (the `command=` value resolution and reading its own module via `__file__` instead of the filename-replace the probe used). No second consistent reading stands, so the step does not return to planning: F29 carries as filed, its next action a later committed brief's text correction of the absorption paragraph.
VERIFIED: Q3 — the report's forms and counts, every figure the gate's own printed line (rule 14), and FINAL is the work commit `fa75ff06c2011745ebf9807dcd6c527e389d9574`, never the report commit `df21eaa` (line 3 of the report; both commits' trailers inspected above). Re-driven at the review tree (test/source-identical to FINAL): `scripts/check-tests --fast` — `983 passed, 109 deselected in 94.67s (0:01:34)`, exit 0 — the expected figures, the report's counts; `.venv/bin/python -m pytest tests/test_batch3_execution.py --durations=5` — `15 passed in 4.04s`, the report's count; `scripts/check-static` — exit 0, gated `All checks passed!`, advisory `Found 104 errors.` — unchanged from REV; `scripts/check-docs` at the FINAL scratch, the hook's offline mode — exit 0, `398 Total … 180 Unique ✅ 395 OK 🚫 0 Errors 👻 3 Excluded` (the three exclusions are lychee-config-excluded URLs, not files), and the gate's online mode prints `398 Total … 180 Unique ✅ 398 OK 🚫 0 Errors` — the report's figures are the online gate's own printed line, faithful. The F28-class baseline arithmetic re-measured at both named revisions: the pristine `cc2ac82` scratch prints `397 Total / 180 Unique / 394 OK / 0 Errors / 3 Excluded` offline, and the documented 397→398 delta is exactly the added brief file (one path in the range, 158 insertions) — the report's corrected arithmetic is accurate at the named revisions. `scripts/check-scope` — `tripped: none`, matching the report's TRIPPED field. The report's isolated two-member line re-driven after the standing battery — `.venv/bin/python -m pytest tests/test_batch3_execution.py::test_every_probe_plan_command_still_names_the_scanned_pair tests/test_batch3_execution.py::test_the_scoped_scan_counts_a_live_pair_holder_and_not_an_argv_carrier -v` → `2 passed in 0.17s`, the report's `2 passed` form. The report's CHANGED line numbers verified against the delivered file (`_PROBE_ARGV` `:36`, `_probe_plan` `:61` with `command=_PROBE_ARGV`, F1 `:248`, F20 `:323`, `_wait_cmdline` `:207`, anchor member `:343` assertion `:358`, conjunction member `:363` assertion `:395`) — all exact. Report fields: every template field present and filled (CHANGED one line per file, VERIFIED with commands and printed lines, NOT VERIFIED filled, FINDINGS in the template's fixed/carried forms, SHAPES, TRIPPED, PROPOSED ROWS, METHOD FRICTION, INHERITED), nothing outside the template added. INHERITED against the received set: the probe-scope review's F27 and F28 blocks are byte-identical in the report's INHERITED (python-diff over the moved blocks); F27's disposition in FINDINGS is fixed at FINAL in the brief's expected rule-8 form with red evidence re-driven above; F28 stays carried with its "not corrected here" claim true of this report; F29 added carried. Received-set dispositions stand as the brief's expected-dispositions line requires (F11/F12/F18/F19/F20/F24/F25 fixed as recorded; F13–F17, F21–F23, F26 carried verbatim with owners).
VERIFIED: standing battery — `scripts/check-tests` (no arguments, the full run, slow modules included, this review's falsification ground): `1092 passed in 1013.25s (0:16:53)` — the counted whole (983 fast pins + 109 slow), all green, no skip and no flake; `tests/test_batch3_execution.py ...............` runs through green (15 pins, both absorbed members included) and `tests/test_phase3_project_oracle.py ......` too (all six oracle drives — the newly unmarked members do not break the oracle's inner deterministic suite).
NOT VERIFIED: the report's live hook line (the work commit's pre-commit hook: parallel phase `949 passed in 22.07s`, serial phase `34 passed in 12.78s`) is not re-driven in that exact form — re-committing would touch the history under review; its class (the full suite green over the anchored module, both phases) is re-driven by this review's own fast run (983 passed, the batch3 module's serial phase included) and the full standing battery above. Everything else the report claims was re-driven or inspected at the named revisions.
FINDINGS: none. The three questions to attack all stand attacked-and-not-falsified: the anchor is one source with the drift class red at the delivered module and at REV; the conjunction member states the selectivity in one scan at FINAL (the `:392`/`:395`/`:398` triple on one `_probe_pids()` call); F29's absorb-both reading is the only one consistent with the rest of the brief. The report's figures are the gate's own printed lines at FINAL and its dispositions match the brief's expected rule-8 forms.
SHAPES: none — every shape this review probed the suite already builds. The anchored-scan shape (F27) is held whole by the committed pair: the anchor half by the one source `_PROBE_ARGV` driven by all three plans plus the committed anchor member (C1/C2 reds re-driven at the delivered module above), the selectivity half by the committed conjunction member (carrier-not-counted / holder-counted / exactly-the-holder, one scan, re-driven above); the drift class's red at REV is the absorbed probes' reds on the REV scratch plus the stacked weakened-capture battery — precisely the reds the brief's red-evidence names, all replayed. Nothing probed is left unbuilt; the deleted probe module held nothing else (verbatim-move check above).
TRIPPED: none
PROPOSED ROWS: none
METHOD FRICTION: none — the scratch-tree `.venv` resolution gap the implementer's report names was applied as the recorded workaround from the start (the shared `.venv` symlinked into both scratch trees before any gate line was trusted); no other deviation from the method or its templates.
INHERITED:

From `docs/research/phase6-probe-scope-implementer.md`, its `INHERITED` block verbatim:

  From the brief's INHERITED block, the items still open, copied verbatim (the chain through the sweep's implementer report; the review's copy moves the same F13–F17/F21/F22/F23 blocks):

  F13 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried
  The brief names the pin's red as "N consecutive grades of the known-good tree produce non-identical verdicts (PASS on one run, CANDIDATE_TIMEOUT on the next)". In this environment the known-good tree's `gate-check-tests` grade is stable well under the fixture's 60.0s deadline (ten pin runs, 50 grades, ~37–46s each, every grade PASS), so the pin passes at REV. The lottery's grade-flip needs a candidate whose suite fails — a failing tree's hypothesis tests shrink, and the shrinking varies the suite's wall — and the near-deadline flip is the neutralisation test's broken-source candidate (FAIL→CANDIDATE_TIMEOUT, 3/3 at REV). The pin's assertion and the seeded determinism are correct; the brief's stated candidate is simply not the one that flips here.
  FIXED: no regression test — an ORIGIN: BRIEF finding is fixed once a later committed brief corrects the text; the pin and the seed land as delivered.

  F14 (P2, ORIGIN: DOCUMENT, carried) — the F8 red evidence is load-borne and the implementer report does not say so; a later actor re-driving the red on an idle host sees green and may conclude the red is chimeric. Next action is the planner's: carry the load condition into the carried F8 block (or a later committed brief's correction), the seed and the pins stand as delivered.

  F15 (P2, ORIGIN: CODE, carried) — killed runs leak intermediate probe trees and no janitor reaps them; the shared tmpfs inode table filled to 3 free and the standing battery's first run died ENOSPC until 81 leaked trees were pruned by hand. Next action is the harness owner's (a tombstone/janitor keyed by the run identity) or a planner ruling that scratch hygiene is each session's duty.

  F16 (P2, ORIGIN: CODE, carried) — the seeded determinism is wall-bounded: at the fixture's 60.0s deadline the near-deadline candidate's seeded grade flips FAIL→CANDIDATE_TIMEOUT under host load (measured once mid-run and once at a 62.44s wall), so the pin's green determinism is stable-half evidence only. Next action is the fixture owner's: raise the grading deadline for near-deadline candidates (the 300.0 default gives a >=5x margin here) or an OWNER ruling that the deadline's load envelope is governed, not deterministic.

  F17 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried
  The brief's boundary-fix instruction is internally inconsistent (FINDINGS, F17): the double-fork pin must leave no descendant alive after the exit while the F10 q1e boundary shape must survive the fix, and the two name one live shape (a session-escaped middleman spawning a session-escaped sleeper then exiting before the deadline, the sleeper adopted by init) that no container boundary can kill and spare at once. The GOAL reading was taken (F11/F12 fixed by the container boundary, both pins green at FINAL, no committed discriminating member changed verdict); the boundary this step retires is the named init-adopted orphan, and the new boundary is what the child never forked. Next action is the planner's: a later committed brief's text correction — the q1e-survival sentence, F12's "without also catching the documented boundary" clause, and the boundary sentence in the module docstring this step corrected in code.
  FIXED: no regression test — the reading is delivered in the work commit and its evidence is the two reds above and the FINAL battery above; the correction itself is a committed brief's prose, which has none.

  F21 | SEVERITY: P2 | ORIGIN: DOCUMENT | STATE: carried
  The implementer report's supervisor-module battery line records the module battery widened by the review module (`.venv/bin/python -m pytest tests/test_phase6_supervisor.py tests/test_review_phase6_supervisor.py --durations=5` — `15 passed in 17.72s`) and never the brief's documented form (`.venv/bin/python -m pytest tests/test_phase6_supervisor.py --durations=5`) with the documented form's own printed line. The wide form is a strict superset (the module's 13 pins plus the review module's 2 probes), green, and every figure it carries is its own printed line — nothing is evaded; but the brief's acceptance names the documented command, and the report's own template requires each claimed figure to name the command that actually ran while this acceptance line names both the documented form it did not run and the wide form it did. This review ran the documented form at FINAL — `13 passed in 12.46s` — so the figures stand; what carries is the report-text deviation from the brief's named battery.
  FIXED: no regression test — a document deviation in a committed report, which only a later step (or the planner's correction of the carried note) absorbs; the documented form's gate line is in this review's VERIFIED.

  F22 (P2, ORIGIN: BRIEF) — the brief's review-block copy of the carried F17 finding is truncated mid-sentence (the literal `(line truncated to 2000 chars)` marker), which rides as a carried text-correction to the planner.

  F23 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried
  The brief's named red at REV is not reproducible — "at REV, with the delivered single-capture _kill, the racing forker's final tail forks plus the double-forked sleeper survive teardown adopted out-of-tree". In a NAMESPACE_V1 drive the workload is the PID-namespace init, _descendants always captures it (descendants=128 namespace_inits=[…]), and its SIGKILL makes the kernel reap the whole namespace (init death kills the namespace; measured: 124 live descendants after a weakened kill, 0 after the init dies), so the delivered _kill already leaves no descendant alive and the pin passes at REV. The sweep lands as the brief's requested defense-in-depth port and the pin as a regression pin; the named red is chimeric for the sandbox, and the load-borne leak F20 measured rode the nested-grading/oracle path, not a standing sandbox-drive race.
  FIXED: no regression test — an ORIGIN: BRIEF finding is fixed once a later committed brief corrects the text; the sweep and the pin land as delivered, and the pin's red class is the kill-surface mutant above.

  Then each finding marked carried by this step, with its ID:

  F26 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried (the repoint's assertion form — the precise inverse `assert carrier.pid not in found` taken over a literal whole-scan `assert not found`; see FINDINGS above; its next action is a later committed brief's text correction).

  Received-set dispositions kept: F19 fixed by the prior step's absorbed pin; F20 fixed as recorded in the sweep report; F11/F12/F18/F25 fixed as recorded in the reports that filed them; F24 fixed by this step (above); F13–F17, F21, F22, F23 carried with their named owners and text corrections.

From `docs/research/phase6-probe-scope-review.md` — its `INHERITED` block, then each carried finding's block from its `FINDINGS` section, moved verbatim (rule 2 governs the planner's copy of this block; the review's received set is the probe-scope brief's carried chain plus its F24/F26, F24 fixed at FINAL, F26 carried):

  From the brief under review, the still-open items copied verbatim (the chain through the probe-scope step's own carried set):

  F13 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried
  The brief names the pin's red as "N consecutive grades of the known-good tree produce non-identical verdicts (PASS on one run, CANDIDATE_TIMEOUT on the next)". In this environment the known-good tree's `gate-check-tests` grade is stable well under the fixture's 60.0s deadline (ten pin runs, 50 grades, ~37–46s each, every grade PASS), so the pin passes at REV. The lottery's grade-flip needs a candidate whose suite fails — a failing tree's hypothesis tests shrink, and the shrinking varies the suite's wall — and the near-deadline flip is the neutralisation test's broken-source candidate (FAIL→CANDIDATE_TIMEOUT, 3/3 at REV). The pin's assertion and the seeded determinism are correct; the brief's stated candidate is simply not the one that flips here.
  FIXED: no regression test — an ORIGIN: BRIEF finding is fixed once a later committed brief corrects the text; the pin and the seed land as delivered.

  F14 (P2, ORIGIN: DOCUMENT, carried) — the F8 red evidence is load-borne and the implementer report does not say so; a later actor re-driving the red on an idle host sees green and may conclude the red is chimeric. Next action is the planner's: carry the load condition into the carried F8 block (or a later committed brief's correction), the seed and the pins stand as delivered.

  F15 (P2, ORIGIN: CODE, carried) — killed runs leak intermediate probe trees and no janitor reaps them; the shared tmpfs inode table filled to 3 free and the standing battery's first run died ENOSPC until 81 leaked trees were pruned by hand. Next action is the harness owner's (a tombstone/janitor keyed by the run identity) or a planner ruling that scratch hygiene is each session's duty.

  F16 (P2, ORIGIN: CODE, carried) — the seeded determinism is wall-bounded: at the fixture's 60.0s deadline the near-deadline candidate's seeded grade flips FAIL→CANDIDATE_TIMEOUT under host load (measured once mid-run and once at a 62.44s wall), so the pin's green determinism is stable-half evidence only. Next action is the fixture owner's: raise the grading deadline for near-deadline candidates (the 300.0 default gives a >=5x margin here) or an OWNER ruling that the deadline's load envelope is governed, not deterministic.

  F17 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried
  The brief's boundary-fix instruction is internally inconsistent (FINDINGS, F17): the double-fork pin must leave no descendant alive after the exit while the F10 q1e boundary shape must survive the fix, and the two name one live shape (a session-escaped middleman spawning a session-escaped sleeper then exiting before the deadline, the sleeper adopted by init) that no container boundary can kill and spare at once. The GOAL reading was taken (F11/F12 fixed by the container boundary, both pins green at FINAL, no committed discriminating member changed verdict); the boundary this step retires is the named init-adopted orphan, and the new boundary is what the child never forked. Next action is the planner's: a later committed brief's text correction — the q1e-survival sentence, F12's "without also catching the documented boundary" clause, and the boundary sentence in the module docstring this step corrected in code.
  FIXED: no regression test — the reading is delivered in the work commit and its evidence is the two reds above and the FINAL battery above; the correction itself is a committed brief's prose, which has none.

  F21 | SEVERITY: P2 | ORIGIN: DOCUMENT | STATE: carried
  The implementer report's supervisor-module battery line records the module battery widened by the review module (`.venv/bin/python -m pytest tests/test_phase6_supervisor.py tests/test_review_phase6_supervisor.py --durations=5` — `15 passed in 17.72s`) and never the brief's documented form (`.venv/bin/python -m pytest tests/test_phase6_supervisor.py --durations=5`) with the documented form's own printed line. The wide form is a strict superset (the module's 13 pins plus the review module's 2 probes), green, and every figure it carries is its own printed line — nothing is evaded; but the brief's acceptance names the documented command, and the report's own template requires each claimed figure to name the command that actually ran while this acceptance line names both the documented form it did not run and the wide form it did. This review ran the documented form at FINAL — `13 passed in 12.46s` — so the figures stand; what carries is the report-text deviation from the brief's named battery.
  FIXED: no regression test — a document deviation in a committed report, which only a later step (or the planner's correction of the carried note) absorbs; the documented form's gate line is in this review's VERIFIED.

  F22 (P2, ORIGIN: BRIEF) — the brief's review-block copy of the carried F17 finding is truncated mid-sentence (the literal `(line truncated to 2000 chars)` marker), which rides as a carried text-correction to the planner.

  F23 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried
  The brief's named red at REV is not reproducible — "at REV, with the delivered single-capture _kill, the racing forker's final tail forks plus the double-forked sleeper survive teardown adopted out-of-tree". In a NAMESPACE_V1 drive the workload is the PID-namespace init, _descendants always captures it (descendants=128 namespace_inits=[…]), and its SIGKILL makes the kernel reap the whole namespace (init death kills the namespace; measured: 124 live descendants after a weakened kill, 0 after the init dies), so the delivered _kill already leaves no descendant alive and the pin passes at REV. The sweep lands as the brief's requested defense-in-depth port and the pin as a regression pin; the named red is chimeric for the sandbox, and the load-borne leak F20 measured rode the nested-grading/oracle path, not a standing sandbox-drive race.
  FIXED: no regression test — an ORIGIN: BRIEF finding is fixed once a later committed brief corrects the text; the sweep and the pin land as delivered, and the pin's red class is the kill-surface mutant above.

  F26 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried
  The repoint instruction names the assertion as "the assertion inverted (the carrier alive, the scoped scan empty)". The probe's old line is `assert carrier.pid in found`; the reading taken is its precise inverse, `assert carrier.pid not in found` (the scoped scan is empty *of the carrier*), with the carrier's liveness asserted beside it. The parenthetical also admits a literal whole-scan `assert not found` reading, which would red on any live workload survivor — the F1/F20 class the brief says stays with the pins — and would couple this host-side probe to unrelated host state, the over-match brittleness F24 is about. Next action is the planner's: a later committed brief's text correction naming the assertion form.
  FIXED: no regression test — an ORIGIN: BRIEF finding is fixed once a later committed brief corrects the text; the repoint lands as delivered (carrier-pid inversion) and is green at FINAL.

  Then each finding marked carried by this step, with its ID:

  F27 | SEVERITY: P2 | ORIGIN: CODE | STATE: carried (the seam's duplicated literal and the buried conjunction member — plan-shape drift silently kills the survivor pins' red class; see FINDINGS above; its next action is a later brief absorbing this review's two probes).

  F28 | SEVERITY: P2 | ORIGIN: DOCUMENT | STATE: carried (the report's docs-baseline claim falsified — pristine `12f1f83` prints 395; see FINDINGS above; its next action is the planner's correction of the carried note).

  F27 | SEVERITY: P2 | ORIGIN: CODE | STATE: carried
  The scoped scan's surface is a duplicated literal: `_PROBE_ARGV` (`tests/test_batch3_execution.py:184`) sits beside three independent plan `command` tuples (`:49`, `:231`, `:306`) and nothing holds them equal, while the named scoped-scan shape is split across the repointed carrier probe (carrier half) and the F1/F20 pins (survivor half) so no committed member asserts selectivity in one scan. Measured: plan-shape drift leaves every committed pin green while the scan undercounts — C1 (a second argv element in all three plan tuples) `13 passed in 4.04s`, C2 (the interpreter path moved to `/usr/bin/python3.13`) `13 passed in 3.90s`; and the drift silently destroys the pins' red class — C1 stacked with the weakened-capture mutant (that mutant alone reddens F1/F20 on `assert survivors == []`) goes `2 passed in 1.64s` over 123 live workload survivors whose argv carried the drifted shape (enumerated and reaped post-run; the pair-matching set empty throughout), leaking past the `finally` cleanups (`:244`, `:319`) that inherit the same blind scoping. The seam is one rule in form (no exclusion list, rule 13) but is keyed to a copy of the surface rather than to the surface the pins grade; the burial of the conjunction member is what lets the anchor drift stay invisible.
  FIXED: the machines this review commits: `tests/test_review_phase6_probe_scope.py::test_every_probe_plan_command_still_names_the_scanned_pair` (AST-parses the batch3 module and asserts every plan `command` tuple equals the scanned pair; green at HEAD `1 passed`, red on C1 with `At index 1 diff: '-u' != '/work/probe.py'` and on C2 the same class) and `tests/test_review_phase6_probe_scope.py::test_the_scoped_scan_counts_a_live_pair_holder_and_not_an_argv_carrier` (one scan over a live carrier and a live holder built from the plan's own `command`: green at HEAD `2 passed in 0.29s` for the module; red on the REV whole-table scan `assert carrier.pid not in found` — `the scoped scan counted an unrelated argv carrier: [3526004, 3526010, 3526011]` — and red on drift when the plan-built holder stops being counted). Next action is the planner's: a later brief absorbs these members into the pinned module (or pins the anchor its own way) — the drift class itself has no single node id at REV, its red is the mutant battery named above.

  F28 | SEVERITY: P2 | ORIGIN: DOCUMENT | STATE: carried
  The report's docs-baseline claim is falsified: "a pristine `12f1f83` extraction prints the same `396 Total`, so the brief's `395` baseline is not this step's" — the pristine `12f1f83` worktree prints `395 Total / 180 Unique / 395 OK / 0 Errors` (the brief's EXPECTED WALLS line, correct at drafting), and the 395→396 delta is exactly `docs/research/phase6-probe-scope-brief.md` added in the reviewed range (docs markdown count 160→161 across `git diff 12f1f83 14b6862 --stat -- docs/`, one path). The figure `396 Total … 0 Errors` is the gate's own printed line at FINAL and stands under rule 14 — what fails is the explanatory claim and its dismissal of a baseline that reproduces.
  FIXED: no regression test — a document deviation in a committed report, which only a later step (or the planner's correction of the carried note) absorbs; the measured correction is in this review's VERIFIED.

Expected dispositions (rule 8 forms): F27 fixed with red evidence (named below — the anchor holds the copies equal or the conjunction member builds the selectivity in one scan, red on the drift class, green at FINAL). F28 stays carried (a document deviation in a committed report, absorbed only by a later step's correction). F11/F12/F18/F19/F20/F24/F25 stay fixed as recorded in the reports that filed them. F13–F17/F21–F23/F26 stay carried with their named owners and text corrections.

Then each finding this review leaves carried, with its ID:

  F28 | SEVERITY: P2 | ORIGIN: DOCUMENT | STATE: carried (the prior report's docs-baseline claim falsified — pristine `12f1f83` prints 395; see the verbatim block above; its next action is the planner's correction of the carried note; this step's own report re-does the delta arithmetic correctly — see this review's VERIFIED).

  F29 | SEVERITY: P2 | ORIGIN: BRIEF | STATE: carried (the probe-absorption vs anchor reading — both members absorbed into the pinned module; see the verbatim FINDINGS block the implementer filed and this review's VERIFIED Q2; its next action is a later committed brief's text correction of the absorption paragraph's "reddens on nothing" clause).
