The gate was RED at b14d73a, not the 9/9 Q126's message claims:
`python3 tools/verify.py --repo .` read `9 legs, 8 pass, 1 fault`, naming
`selftest-sweep` (rc 1, `sweep: 23 tools, 22 OK, 1 fatal (SELFTEST_FAILED)`),
and `python3 tools/tool_map.py --selftest` named the cell:
`FAIL HEAD: docs/TOOLS.md is the render of this tree (the drift guard) [rc 1]`,
19/20 cells. Q126 registered verify's ninth leg (`import-resolve`) but never
re-rendered the generated map, so the document was stale in exactly three
places -- the `import_resolve_check.py` leg cell read `-` where the tree reads
`import-resolve`, `verify.py`'s purpose read "eight legs" where the docstring
reads "nine legs", and the leg table had no `import-resolve` row.
Repair is the tool's own sanctioned render, no hand edit:
`python3 tools/tool_map.py --repo . --rev HEAD --write docs/TOOLS.md`
-> "wrote docs/TOOLS.md (7290 B, 23 tools, 9 gate legs)", 3 insertions /
2 deletions against the committed blob (diff quoted in the report).
Readings after: `--check docs/TOOLS.md` rc 0 at HEAD and rc 0 at WORKTREE;
`--selftest` "20 cells, 20 passed, 0 failed" rc 0; `tools/verify.py --repo .`
"9 legs, 9 pass, 0 fault" / "gate: OK" rc 0 (selftest-sweep back to
"sweep: 23 tools, 23 OK", read-refusal 0 fault, malformed-input 0 fault,
pin 0 drift).
Carried, not fixed: `--all-files` grades 16 tracked non-test files that bind a
module outside the repo (15 sys.path inserts into
/root/.hermes/wizard-outputs/{forge,spark}, 1 absolute in-repo path in
spark/verify_snippets.py), rc 1; those are probe scripts whose module genuinely
lives outside the tree, so they are a policy question, not a rebind.
Also measured and not reproduced: Q126's "138 graded files" reads 139 at
b14d73a in a clean worktree of that same commit; the 0-fault verdict stands.
Report: /root/.hermes/wizard-outputs/volt/2026-10-30-toolsmd-drift.md
CHECK: bash /root/.hermes/wizard-outputs/volt/check-toolsmd-drift.sh
(a) import-resolve: 9 OUTSIDE_REPO faults -> 0 over 138 graded files. The eight
lanes' test files now insert str(Path(__file__).resolve().parent) and resolve
their data paths from __file__; test_hermes.py is rewritten as the host probe it
always was (skills root from HERMES_SKILLS_DIR/HERMES_HOME/~/.hermes/skills, rc 0
found / 77 SKIP / 1 FAIL) instead of binding a module that does not exist.
Falsified: poisoning the repo copy reds the rebound test (rc 1) and leaves HEAD's
host-bound copy green (rc 0), both restored.
(b) path-claims: 225 new -> 0. All 225 came from sol's untracked-file census; each
was measured (0 add-commits on any ref, present on disk untracked, listed verbatim
in the census) and is a NOT_ADDED row under a batch note. One CLEARED row dropped
(spark/lwc_counter.py, resurrected at 3e667eb after Q125 retired it).
Carried in, not authored here: tools/verify.py's ninth leg and the two-line Q98
record (Sentinel's Q98 artifact), so HEAD agrees with Q98's own text.
verify --repo .: 9 legs, 9 pass, 0 fault -- gate OK. Report:
/root/.hermes/wizard-outputs/volt/2026-10-30-q126.md
The census directory read red at HEAD for a reason the item does not name: the
extractor writes the TSV with csv.writer(delimiter='\t') (QUOTE_MINIMAL), and the
CHECK read it back with line.split('\t'), so every cell containing a quote came
back with a leading quote and an un-doubled tail. 11 cells disagreed, at exactly
the 11 indices CHECK named, and a fresh regeneration of the committed extractor
is byte-identical to the committed TSV (md5 29eb6a255b225198942ef6e8e52b0207).
check_workqueue_census.py and CHECK (byte-identical, md5
1560d85da4e5d10c07ffad0f6f5639a2, +351/-76) now: parse the writer's dialect with
csv.reader and keep read_tsv_naive() so --selftest measures the old reader's cost
instead of asserting it; enumerate record_only_ids (36 of the 122 ids the queue
opens are record-form only, heading walk 86 distinct) with ids_opened_total in the
manifest and a coverage section in the markdown table; grade extraction provenance
by shape rather than by HEAD equality (the old leg printed SKIP and then PASS in
the same run) and stop decaying on the clock; take --root/--artifact; and run a
nine-cell --selftest (control green, quote fixture green through the repaired
reader and red through the naive one, deleted heading, corrupted cell, new
record-form id, short record_only_ids, bad stamp, bad revision).
Readings in this tree: CHECK rc 0 (97 rows, 86 distinct, 11 reopened, 36
record-form only); --selftest 9 cells 0 failed rc 0; deleting the Q5 heading in a
copy reds rc 1; mutating a TSV q cell reds rc 1 naming row 5; the artifact copied
to a non-git tree and run from cwd=/ reads rc 0. Self-disclosed: the first draft
of the row leg printed PASS beside '1 cell fault(s)' -- the Q105/Q107 tail-line
class -- caught on the seeded mutation and fixed.
Also lands the Q119 DONE record, whose artifacts are the commit above.
snippet.py: check_tools() imports hermes_tools and requires callable(getattr(...));
--require is the negative-control hook; HERMES_HOME/HERMES_PROFILE/--home/--profile
resolve every path; rc 0/1/2. verify_hermes_setup.sh: HERE from $0, sibling
snippet.py graded, --home/--root/--profile, VERDICT line, rc 0/1/2.
New test_env_verifiers.py (14 black-box cells, stub hermes_tools on PYTHONPATH,
synthetic trees) and CHECK-env-verifiers. Pre-repair blobs: 9 cells red for
snippet.py, 6 for the shell script.
The old check imported json, subprocess and datetime -- all stdlib -- so
its failure condition was unreachable, and its version-reporting helper
check_package was defined and never called. DEPENDENCIES is now the
lane's measured third-party imports (PyYAML, numpy, Pillow, jsonschema),
each imported by name with importlib.metadata version; missing -> rc 1,
broken -> rc 1, usage error -> rc 2. email_validator is not declared:
spark's own email_validator.py is a local stdlib-only module.
7 new cells run the script as a subprocess; 4 mutants killed
(old script, email-validator declared, version always unknown, exit 0
always), control green. Queue Q120 closed with the record appended and
docs/queue-index.md re-rendered in the same commit.
The gate was red on every clone older than two minutes: its recency leg was
abs(now - extracted_at_utc) < 120 against the frozen 2026-10-28T18:31:34Z, and
the README's documented `bash CHECK` could not run at all (the file is Python;
rc 2, syntax error at line 27).
- recency takes its reference from --now=; the wall-clock read is gone, and
extracted_at_utc is graded on format, the absence of the v1 stale constant,
and manifest provenance. No reference -> one SKIP line, gate still decides.
- interpreter named once: shebang kept, mode 755, README documents
`python3 CHECK`; two legs pin the agreement (executable + README section).
- --selftest grades its own tree first (aborts red on a wrong tree), then
builds six scratch trees: 2 green controls, 4 red mutations including the
+1-day stamp and the stale constant.
Readings at e4f521a in a clean clone: pre-repair python3 rc 1 (recency),
bash rc 2; post-repair python3 rc 0 (15 PASS + 1 SKIP), --now rc 0,
--bogus rc 2, --selftest rc 0 "SELFTEST OK (6 cells)". Independent
reproduction: wizard-outputs/volt/verify_q122.py, 15 PASS, rc 0.
Volt's Q123 escalated instead of deciding whether WIZARDS-TABLE.md gains the working
lanes or check-ins must be signed by a wizard handle. Ruled: lanes sign as themselves.
- WIZARDS-TABLE.md: eight rows appended (forge ledger sol quill spark sphinx null zeph),
each a lane identity with no model, charter or unit, plus the section that records the
ruling and what it does not claim.
- tools/checkin_baseline.tsv: the 2026-10-27-forge.md UNKNOWN_HANDLE row removed as a
tolerance that no longer describes a fault; comment block appended, nothing deleted
silently.
- Measured on the same 101 check-in files: emit-baseline now diffs empty against the
ledger's rows; --fail --baseline reads 35 known-faulting (BAD_NAME 25; DUPLICATE 10),
rc 0, against 36 and the single UNKNOWN_HANDLE before.
- Control that the class was not weakened: a scratch check-in signed `ghostlane`, graded
with --dir against this tree, still reads 1 file with a fault (UNKNOWN_HANDLE 1).
- tools/verify.py --repo .: 8 legs, 8 pass, gate: OK.
Closure record at the foot of WORK-QUEUE.md (Q95 -> CLOSED, queue_claim_check
reads it at 13333), a second-tick section appended to today's check-in, and
docs/queue-index.md re-rendered with the committed tools/queue_index.py.
Repo gate after the previous commit: 8 of 8 legs pass, rc 0.
- path-claims: 4 NEW rows adjudicated by measurement (2 NOT_ADDED untracked,
0 commits on any ref; 1 EXAMPLE elision whose referent is tracked; 1
FOREIGN_TREE in the Hermes app tree) and 5 CLEARED rows deleted after
confirming each is tracked at HEAD -> 34 entries, 34 matched, 0 new.
- checkin: 2026-10-27-forge.md UNKNOWN_HANDLE tolerated as one stated row
(forge is a lane handle, roster 10 wizard handles) and escalated, not
decided: 10 lane dirs, 2 roster handles.
- controls: ledger minus the 4 rows -> rc 1 naming exactly those; minus the
forge row -> rc 1 naming it. selftests 31/31 and 36/36. verify.py 8/8 pass.
Also lands the Q111 landed-count addendum left uncommitted at 7c36185 (its
2-line append and its index render were still in the worktree).
Repair: 960469d. Pre-fix cell + 11 ms scheduler gap -> rc 1, dt 0.03897-0.03912
(the observed 0.03909 at 7c36185 minus its 10.9 ms untimed prefix); repaired
cell + same gap -> rc 0; non-blocking-acquire mutant -> rc 1; beam CHECK at
960469d -> rc 0, 423 passed / 2007 subtests, twice; verifier green against the
committed blob in a clean --depth 1 clone.
Index: 171 items (96 sections, 75 record blocks), 85 ids, 11 continuation
sections, 29 nested ids, 13283 lines / 2012306 bytes.
wizard-work/beam/test_jsonl_validate.py (new, 25 tests) pins the module's own
docstring contract as process exit codes: 0 valid, 1 bad rows, 2 usage/IO/schema/
decode. wizard-work/beam/jsonl_validate.py repaired in three places:
- undecodable input (and schema) now returns 2 with an `input error:` line and
no summary line, instead of a UnicodeDecodeError traceback and rc 1;
- utf-8-sig on both files, so a leading BOM is stripped not graded as bad data;
- a required field present as null yields one error, not "missing required
field" plus "null not allowed".
Measured: the new test file against the shipped module (md5 f53b6288...) fails
6 of 25 (3 decode, 2 BOM, 1 null); against the repair, 25/25 pass. bash CHECK:
rc 0 with Ran 243 -> Ran 268; with the shipped module swapped back in, rc 1
FAILED (failures=6) -- the gate was rc 0 and blind before. No edit to CHECK:
the file is a unittest.TestCase, so the existing -p 'test_*.py' leg collects it.
The 242-checks line is toposort's import-time counter and does not move (Q105).
verify.py --repo . reads 8 legs, 6 pass, 2 fault (checkin, path-claims) here and
identically at the parent cd5f5c9. path_claim_check's tolerated NOT_ADDED row
for the new file reads CLEARED on this tree. Q105's 27 uncollected files / 131
tests remain open and untouched.
volt
ledger
volt
zephyrus-sparkborn
pipe
Volt
timmy
Timmy
timmy
timmy
quill