Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 16 additions & 0 deletions BASELINE.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,22 @@ This ledger binds the currently observed repository snapshot to the twelve
owner criteria and final gate matrix. It records evidence and gaps; it is not a
completion certificate.

## Integrated origin and vector observation — 9 November 2026

Main #789 and Workspace #60 were integrated with exact-head checks. Workspace
`defb18ed4745432d3a2f1bc7a17cc28a1292a334` pins Main
`b0a6c12abe87ae2a43163b3fe6ba4fa14adf389d`; these are sequentially observed
parents of this later documentation work, not a permanent final vector.
[STATUS](STATUS.md#later-integrated-origin-and-source-cut-observations--9-november-2026)
records the bounded hosted mutation and regression evidence, offline structural
cut, failed private Git transports, ref inventory and exact limitations. The
[machine-readable ref observation](cbz/inventory/federation-refs-20261109.json)
is historical and does not decide whether a non-ancestor is disposable. A
later Main merge requires a new Workspace pin comparison. Full G0–G11, all
twelve criteria and STATUS-only cold-start acceptance remain unawarded.
Math #559 subsequently merged as `9bdfc54b76173b57912503bfcf901a5f9d43b09c`;
Workspace's earlier Math pin is stale and needs a guarded update.

## Node and pin observation before this change — 9 November 2026

Main `1902aac113761974c4569f78acb6a8baadabb7f1` and Workspace
Expand Down
7 changes: 7 additions & 0 deletions DECISIONS.md
Original file line number Diff line number Diff line change
Expand Up @@ -140,3 +140,10 @@ Append-only. Historical source outcomes remain in their original records.
| 2026-11-09 21:38 | 2 | B/tooling, testing | The expanded inventory adverse-driver attempt failed because its isolated fixture lacked the newly required controller and inventory source files. Retain that failure, copy those exact files into the fixture, and rerun the no-op and subject-origin probes in both parent modes. The preceding local inventory run completed 83 tests per mode without top-level skips; four intentional inner recursion skips remain explicit. These bounded results do not establish a final federation gate. | Yes; failed and corrected outputs retained | Candidate native `87541cd60ef4a065bd3a7a00497c88a894787c4e`; `cbz/evidence/inventory-origin-20261109/run_driver_adverse.py`. |
| 2026-11-09 21:43 | 2 | B/tooling, conflict | Preserve the real freshness-grader false kill where a selected test changed its failureException and raised ValueError. Require the pre-cleanup exception to be exactly AssertionError and verify the report type, with fresh adverse and mutation controls. Earlier results are retained at their actual source boundary and are not rebound. | Yes; red and focused corrected observations retained | Main #789 prior head `aea190621ab5ef0d2c1cfd581bc088fa904a22eb`; main#738 comment `6752146963`. |
| 2026-11-09 21:57 | 2 | A8/B, tooling | Record the distinct Math code-object equality false kills and the separate identity-based Math #559 candidate. Retain both the concurrent-copy failure and the ambiguous nested-transcript self-mutant; require an isolated complete successor matrix and current hosted qualification. The preserved-byte artifact's complete hidden capture is byte-integrity evidence only. | Yes; prior outputs retained | [Math #559](https://github.com/d6g8k5htny-coder/Math-/pull/559); [coordination](https://github.com/d6g8k5htny-coder/main/pull/738#issuecomment-6752496413); [STATUS](STATUS.md#later-node-inventory-and-pin-observations--9-november-2026). |
| 2026-11-09 22:20 | 3 | B/integration, protection | Integrate Main #789 after its exact tested-tree crawler, Node and other hosted checks. Preserve the initial GitHub rejection while required `public-intake` was in progress; retry only after actual success, with the same expected head and no bypass. The current component observations do not award whole-federation G0/G6 or G5. | Yes; original tree and rejection retained | Main merge `b0a6c12abe87ae2a43163b3fe6ba4fa14adf389d`; tested merge `de9db13d21b53ab33dda09818e01d53979d90f8a`; runs `42707612430` and `42707612474`. |
| 2026-11-09 22:33 | 3 | B/tooling, integration | Preserve the vector grader's real spoofed-path false kill and repair its pre-cleanup assertion provenance using actual code identity. Retain the original source mutations, add a real guard-removal self-mutation and refresh the Main pin. Integrate Workspace #60 after its hosted 254-test-per-mode contract and strict pin/canonical checks; arbitrary helpers and hostile runtime remain outside the grader's supported domain. | Yes; prior source and outcomes retained | Workspace merge `defb18ed4745432d3a2f1bc7a17cc28a1292a334`; tested merge `1f5ae4e0c507ee2133da8dc03f16f4e0a8da63b7`; runs `42712020137` and `42712020429`. |
| 2026-11-09 22:36 | 2 | B/inventory, custody | Record the twelve-source offline structural cut and separate before/after default observations without claiming atomicity, full working parity or final G7. Retain the earlier missing-object failure and the network run's two private transport failures. Preserve all observed tags: non-ancestry alone does not establish that a tag is unreferenced or safe to remove, and the explicit quarantine/archive tags carry historical source boundaries. | Yes; original observations and refs retained | Offline receipt SHA-256 `d5a4672c7a07835a72d4edd68ddbd670e9b0bd24f8124b601359c3ab3619ddd5`; earlier network receipt SHA-256 `7aadc625ac39f54cd3b4c29a5e8e4690f4008f5d038967b4ea67642f5068a8ee`; [STATUS](STATUS.md#later-integrated-origin-and-source-cut-observations--9-november-2026). |
| 2026-11-09 22:36 | 2 | A8/B, evidence | Record Math #559's actual current kernel and preserved-byte observations without rebinding historical receipts or asserting continuum/scientific acceptance. The two independently observed verifier-script mistakes were retained and corrected; full repository regression was still running at this decision point. | Yes; prior and corrected evidence retained | Tested merge `79ad170de796404f6a8cd5273ce0fc4229da361e`; kernel receipt SHA-256 `12c4985fdb146abbb500ac39a2874e478b863f257a0b966bc2d030aad4d849d7`; [Math #559](https://github.com/d6g8k5htny-coder/Math-/pull/559). |
| 2026-11-09 22:41 | 2 | B/inventory, privacy | Publish a bounded machine-readable projection of the captured federation refs: all public branch and tag identities with source-bound ancestry observations, plus private aggregate counts. Withhold private branch names and commits from the public projection; retain the complete private observations in the owner's evidence. Native traversal covered ten public default histories, and authenticated API comparisons covered the five private branch tips. Ancestry is not semantic merge, custody, deletion eligibility, full Git hygiene or final G10. | Yes; prior refs and observations retained | [Ref observation](cbz/inventory/federation-refs-20261109.json); 2,078 captured branches, 39 tags; private Git transports previously failed. |
| 2026-11-09 22:43 | 2 | A5/B, evidence | Preserve current observed failures on the separate Main #787 and #788 drafts, including both-mode component controls and normal-discovery failures and skips. Do not count those failed runs as qualification or change their scopes in this integration draft. | Yes; original artifacts retained | #787 head `d561405d936e81d2f77e9308b4e176cb994c77aa`, artifact `13715491858` / normal discovery `13714354922`; #788 head `f5f301b652e222395dc45cf300ae6697c4fcf8ae`, artifact `13714880889` / normal discovery `13715662136`. |
| 2026-11-09 22:45 | 3 | A8/B, integration | Integrate Math #559 after exact current-head kernel, preserved-byte, downstream and full-regression workflows succeeded. Its full Math suite ran 1,144 tests per mode with no skips; original statements, seven actual negative controls and historical receipts retain their source identities. No continuum or scientific acceptance is inferred. Workspace's previous Math pin must be refreshed through the normal guarded path. | Yes; premerge source and failures retained | Math merge `9bdfc54b76173b57912503bfcf901a5f9d43b09c`; tested merge `79ad170de796404f6a8cd5273ce0fc4229da361e`; kernel run `42708585053`; regression run `42708585331`. |
15 changes: 15 additions & 0 deletions GOVERNANCE.md
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,21 @@ Platform permissions and denials remain binding. Historical mathematical
statements, test failures, authorship and verification outcomes are facts;
this governance change does not rewrite them.

## Later integrated provenance boundary — 9 November 2026

Main #789 and Workspace #60 integrated the bounded mutation-origin fixes and
the then-current Main pin by ordinary platform-enforced merges. Workspace's
vector mutation grader admits only the selected synchronous unittest bodies
and its explicitly recognized standard unittest assertion helpers. Its
commands, adverse controls, observed source cut and limitations are recorded
in [TOOLING.md](TOOLING.md#workspace-vector-and-pin-checks--9-november-2026)
and [STATUS.md](STATUS.md#later-integrated-origin-and-source-cut-observations--9-november-2026).
An earlier two-private-transport failure is retained; API-reconstructed local
Git objects do not qualify as successful remote Git fetches. No component
result substitutes for the final same-pass federation gates.
Math #559 has since merged after its own current kernel, byte and full-suite
checks; Workspace's earlier Math pin needs a fresh guarded update.

## Current owner source

The latest complete master is preserved [verbatim in this repository](governance/MASTER_DIRECTIVE_20261029_SWARM.txt)
Expand Down
21 changes: 21 additions & 0 deletions PROOF_INDEX.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,27 @@ into a kernel proof. A theorem is CLOSED only for the literal statement and
allowed-axiom boundary actually checked. Anything else must remain OPEN or be
placed in a complete dormant quarantine.

## Later source and custody boundary — 9 November 2026

Main #789 and Workspace #60 are engineering and pin changes; they do not
alter mathematical statements, bounds, authorship, proof outcomes or
quarantine dispositions. The separate Math #559 tested merge
`79ad170de796404f6a8cd5273ce0fc4229da361e` produced a kernel receipt for
66 exact conditional targets and seven actual rejection controls, with
axioms limited to `Classical.choice`, `Quot.sound` and `propext`. The
original 66 statements, 78 subject bodies and historical receipts remain
at their own source identities. No continuum parent claim or independent
scientific acceptance follows. Math #559 subsequently merged as
`9bdfc54b76173b57912503bfcf901a5f9d43b09c` after 1,144 repository tests
passed in each mode without skips; this does not establish federation G5.

Eight non-ancestor tags observed in the current bounded ref inventory are
explicit preservation/quarantine snapshots. Main's three historical tags
are referenced by [the cold-start guide](docs/cbz/COLD_START.md); five Math
tags preserve the #535 carrier's original histories. Their existence is not
itself proof of the eleven-field quarantine contract, logical isolation or
kernel closure. Exact observations and limits are in STATUS.

## Later engineering boundary — 9 November 2026

Main #785 and #784, Workspace #58, the prepared Node regression and the later
Expand Down
107 changes: 107 additions & 0 deletions STATUS.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,113 @@ The inherited scientific material below is explicitly the 29 September
snapshot, not a fresh CLOSED decision or an empirical-validation receipt.
Its mathematical statements, source cuts and review limits are preserved.

## Later integrated origin and source-cut observations — 9 November 2026

Main #789 merged as `b0a6c12abe87ae2a43163b3fe6ba4fa14adf389d` and
Workspace #60 merged as `defb18ed4745432d3a2f1bc7a17cc28a1292a334`;
the latter pins that Main commit. Math default was separately observed at
`f532c3e2d91a0c7679a37c733741660917a9b21b` before #559 merged as
`9bdfc54b76173b57912503bfcf901a5f9d43b09c`; Workspace's Math pin therefore
requires refresh. These observations predate this documentation composition:
a later Main merge also requires a fresh guarded Workspace pin comparison.
The first #789 merge request was rejected while an
actual required `public-intake` check was in progress; after that check
succeeded, the normal expected-head retry merged without bypass.

At Main #789's tested merge `de9db13d21b53ab33dda09818e01d53979d90f8a`,
the hosted crawler ran 50 frozen controls and 23 grader controls per mode,
and its six shards and merges reported 109/109 designated source mutations
in each mode. The hosted prepared Node run had 33 helper controls and 767
passing Node tests per mode without skips, cancellations or todo cases.
Original artifact hashes, source identities, crawler provenance and Node TAP
and 39 fixture bytes per mode were independently checked. These are bounded
component results, not a no-skip whole-federation regression or final G0/G6.

Workspace #60 repairs a reproduced vector mutation false kill caused by a
subject under a directory named `unittest`. Its grader uses pre-cleanup live
code identity for the selected body, dispatch and standard unittest helpers;
arbitrary helpers and hostile runtime manipulation remain unsupported. At
tested merge `1f5ae4e0c507ee2133da8dc03f16f4e0a8da63b7`, hosted contract
run `42712020137` ran 254 tests per mode with no skips and recorded 13 clean
baselines and 13 designated source mutants per mode. Pin run `42712020429`
checked eight strict-current pins and four canonical inputs. The original
artifact and all 52 nonce/source/stderr-bound events were checked. These are
sequential observations, not a final G7 or G5.

A new offline vector run on a separately observed twelve-repository cut at
those Main, Math and Workspace commits passed its selected structural checks:
12 commits, 30 selected inputs, eight pointers, four canonical bindings and
nine lock packages. Receipt SHA-256
`d5a4672c7a07835a72d4edd68ddbd670e9b0bd24f8124b601359c3ab3619ddd5`.
Separate before/after API reads matched all twelve heads; they were not atomic.
The earlier cut first failed on a missing historical Main Git blob; after
recovering and verifying that exact object it passed offline. Its network run
fetched 20 of 22 exact commits and failed because two private Git transports
returned exit 128. No current private Git transport success, full tracked
working parity or Git hygiene is inferred.

The bounded remote inventory observed 2,078 branches and 39 tag refs across
the twelve visible repositories. Ancestry comparisons found 1,441 branch
tips ancestral to the captured defaults and 637 non-ancestral; neither result
decides semantic integration or deletion eligibility. The
[machine-readable observation](cbz/inventory/federation-refs-20261109.json)
records public refs and private aggregate counts; private branch names and
commit IDs are withheld from the public projection. Twenty-nine tag targets were ancestors of
the captured defaults; ten were not. Eight non-ancestors are explicitly
named preservation/quarantine snapshots and remain retained; non-ancestry
does not establish that any tag is unreferenced or safe to remove. The local
cut's clean statuses and empty stash lists cover only those checkouts and their
registered worktrees, not every clone or orphan worktree.

Math #559 merged as `9bdfc54b76173b57912503bfcf901a5f9d43b09c` after
tested merge `79ad170de796404f6a8cd5273ce0fc4229da361e` completed its
kernel and byte-integrity workflows. Its receipt records 66
literal target axiom lists, seven real rejection controls and no scientific
effect; the original conditional statements and prior outcomes are not
rebound. Full-regression run `42708585331` completed 1,144 normal and 1,144
optimized Math tests with no skips. This is not a full federation G5.
Neither the parent continuum claims nor independent scientific acceptance
follows.

At the separately observed Main #787 head `d561405d936e81d2f77e9308b4e176cb994c77aa`,
the hosted scope contract ran 23 tests per mode and failed two
submicrosecond-fraction controls in each. Its normal discovery ran 2,599
tests with two failures and 25 skips. At Main #788 head
`f5f301b652e222395dc45cf300ae6697c4fcf8ae`, the hosted offline contract
ran 23 tests per mode successfully, while the 18 CLI tests had 33 failures
in each mode; normal discovery ran 2,617 tests with 33 failures and 25 skips.
These are observations at those exact draft heads, not results for later
changes. Both deliveries remain separate open scopes; the original failure
artifacts are retained and neither is credited as a completed gate.

From a clean Workspace checkout, the bounded vector commands are:

```sh
python3 -B -S -m unittest discover -s tests -p 'test_verify_federation_vector*.py' -v
python3 -B -O -S -m unittest discover -s tests -p 'test_verify_federation_vector*.py' -v
python3 -B -S tools/verify_federation_vector.py --root /absolute/siblings --vector /outside/vector.json --output /outside/new-offline.json
python3 -B -S tools/verify_federation_vector.py --root /absolute/siblings --vector /outside/vector.json --network --output /outside/new-network.json
python3 -B -S tools/verify_pins.py --require-current
python3 -B -S tools/verify_canonical_inputs.py
```

For the vector commands, place twelve clean Git checkouts directly below the
absolute sibling root. The JSON file must have exactly `schema` set to
`cbz.federation-vector.v1` and `repositories` mapping these exact names to
their selected full lowercase 40-hex commit IDs: `main`, `Math-`,
`Universal-Law-Workspace`, `d6g8k5htny-coder`, `google-drive`, `governance-`,
`meta-framework`, `openclaw-custodian-validation`, `query-`,
`research-hosted-qualification`, `sandbox` and `trial`. The output must be a
new nonsymlink file outside all checkouts. Historical canonical objects must
also be locally present; an unavailable object or private remote is a real
failure, not a skip.

The vector is an explicitly selected immutable cut, not a default-tip or
scientific certificate. Full proof/quarantine reconciliation, complete Layer
A audit, no-skip regression, authenticated private transport, all final
G0–G11 gates and the STATUS-only cold start remain open. No CBZ-0 or terminal
Partial Baseline is declared.

## Later Node, inventory and pin observations — 9 November 2026

Before this source change, Main #785 merged as
Expand Down
31 changes: 31 additions & 0 deletions TOOLING.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,37 @@ are preserved in their dated sections as history. Each entry states purpose,
command, verification and scope. A component passing in isolation does not
complete a CBZ gate or the final G0–G11 matrix.

## Workspace vector and pin checks — 9 November 2026

**Purpose.** Workspace `tools/verify_federation_vector.py` checks a selected
twelve-repository immutable source cut, its eight Workspace pointers, four
canonical bindings and nine lock packages. `tools/verify_pins.py --require-current`
separately compares the eight public pins with observed
remote defaults. `tools/verify_canonical_inputs.py` checks the four selected
canonical source blobs. None proves mathematical statements or an atomic
federation state.

**Commands.** From a clean Workspace checkout, run:

```sh
python3 -B -S -m unittest discover -s tests -p 'test_verify_federation_vector*.py' -v
python3 -B -O -S -m unittest discover -s tests -p 'test_verify_federation_vector*.py' -v
python3 -B -S tools/verify_federation_vector.py --root /absolute/siblings --vector /outside/vector.json --output /outside/new-offline.json
python3 -B -S tools/verify_federation_vector.py --root /absolute/siblings --vector /outside/vector.json --network --output /outside/new-network.json
python3 -B -S tools/verify_pins.py --require-current
python3 -B -S tools/verify_canonical_inputs.py
```

**Verification and scope.** Workspace #60's exact tested merge ran 254 tests
per mode without skips and 13 actual source mutants with clean baselines per
mode. Its adverse spoofed-path subject is refused, and removing the new guard
reopens the false kill in both child modes. The grader uses live pre-cleanup
code identities and supports selected synchronous unittest bodies and standard
unittest helpers, not arbitrary helpers or hostile runtime code. Earlier
filename-based results remain historical. The later offline and earlier
network observations and their exact boundaries are recorded in STATUS;
full tracked working parity and Git hygiene were not assessed.

## Later provenance and prepared Node tools — 9 November 2026

**Inventory and freshness mutation provenance.** The respective tools are
Expand Down
Loading
Loading