Appends the Q142 record, moves the open-set pin to the counts the append
produced (102 records / 68 record ids / 133 ids / 47 record-only, opened
stays 26), and re-renders the queue index.
Leg 4 built its copy from the worktree, so a dirty tree broke the leg that
exists to prove a copy grades itself. It now extracts wizard-work/quill from
HEAD with git archive; legs 2 and 3 remain the worktree readings, and the
script sets pipefail so a failed archive cannot read as an empty tarball.
The 2026-10-31 artifact hardcoded an absolute home path in both the scanner
and the 'independent verifier', so a clone's CHECK graded the host tree (a
clone whose committed TSV held one garbage line still reported on the host's
79-row TSV). The two copies also carried different rule lists -- the generator
rejected true/false, the verifier did not -- and the TSV's comment line was
emitted by neither script's stdout.
- draft_compliance_rule.py: the bare-command rule, stated once, 11 cases
- generator: paths from __file__, artifact excluded from its own population,
deterministic output (comment + header + rows)
- verifier: own re-derivation, both directions, grades the report's six counts,
rc 0/1/2, --selftest one control + eight mutants + rule-agreement + no-host-path
- CHECK: four legs (selftest, rederive, byte-identical regeneration, copy-parity)
- report: measured numbers (79 drafts, 14/14/11, 68 failing, 3 non-bare)
verify_line_counter.sh pinned the pre-Q125 tool at a host path
(/root/.hermes/wizard-outputs/spark/line_counter.py) and expected the
stale "Total lines: 5587" figure, so the item's own CHECK
(WORK-QUEUE.md:13377) reddened: usage-line output, rc 1.
Repair (confined to wizard-work/spark):
- verify_line_counter.sh resolves line_counter.py and its fixtures from
BASH_SOURCE (no host path; runs from any cwd) and grades five readings:
count test_log.txt == wc -l == 10; csv test_data.csv == csv-module
field count == 5; count test_blank.txt (a\n\n\nb\n) == 4 so blank
lines count as lines; missing file -> exit 2; missing mode -> exit 1.
- test_line_counter.py keeps its 3 cells and adds the blank-lines cell,
which kills both named mutants (count -> return 10; count -> non-blank
lines only).
- test_blank.txt: new fixture, exactly printf 'a\n\n\nb\n' (4 lines).
- README.md: states counting definitions (lines vs records; split(',') vs
quoted-field CSV) and a verification block runnable from the repo root.
- VERIFICATION.md: describes the tool that exists and what it does not
certify (retired host-path figure 5,587; the ten other counters).
Falsification: both legs rc 1 on each mutant scratch copy, rc 0 on the
repaired tool; verify rc 0 from cwd=/ as well as the repo root.
Q125's own CHECK exits 0 at this HEAD.
docs/TOOLS.md drifted on two counts (ten legs, missing open-set row); tool_map.py --write now renders 23 tools / 11 gate legs. open_set.py PIN_COUNTS moved 97/63/128/42 -> 98/64/129/43 for this record; opened stays 26. Full gate 11/11 at the committed tree.
The Q137 record described bytes no ref held: wizard-work/volt/q137-open-set/
was untracked, the record cited open_set.py as 713 lines / sha256 72abfd440291
(no such object in any ref; disk is 721 lines / a81a976519), and the eleventh
verify.py leg was uncommitted. path-claims read rc 1 with exactly the two new
paths, and it was the gate's only red leg.
Landing: open_set.py, CHECK-q137, REPORT.md, the queue addendum, the
re-rendered index, tools/verify.py's open-set leg.
After: CHECK-q137 4/4 rc 0; verify.py --selftest 59/59; verify.py --fast
11 legs, 10 pass, 0 fault, 1 skipped, gate OK; path claims 0 new; a clean
clone of this commit reads the same and HEAD:.../open_set.py sha256 equals
the digest the addendum states.
sol
volt
ledger
quill
beam
volt
Volt
zephyrus-sparkborn