Commit graph

997 commits

Author SHA1 Message Date
igneum-labs
c811b35a12 Merge master e49bcb841 into registry-ux06 under the master-landing lock 2026-10-08 21:21:34 +00:00
igneum-labs
d9b183f595 Merge master bd8d6a25a into census-rows-live under the master-landing lock 2026-10-08 21:20:16 +00:00
igneum-labs
fb0e6f5eda Merge master 5e9497e7e into class-v6-kits under the master-landing lock 2026-10-08 21:12:43 +00:00
igneum-labs
257b7c310f Class v6 census: the live-dataset census on the frozen object 21edd23c9 (Phase 1 item (j)): hl-v6-all 4de7b836cc40a4ea and sixteen chain seeds on the chain's dataset with the node1 leaves under header-bound init words, uniform inside every window to the Poisson floor, F8 form 0.999x of the window model; F06's live half LIVE on every register and load; the registry run census-v6-live-20261008-2210
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:12:34 +00:00
igneum-labs
55af58de28 Merge master 07105d4dc into ci-guest-format under the master-landing lock 2026-10-08 21:11:41 +00:00
igneum-labs
f42be36506 Merge master e85fad3d9 into d72ea2a3e2 under the master-landing lock 2026-10-08 21:10:37 +00:00
igneum-labs
3200b22dce Merge master e0f7e6762 into class-v6-floor-k-docs-2 under the master-landing lock 2026-10-08 21:09:34 +00:00
igneum-labs
4c0bcc89cf Test registry: UX-06 FAIL on the over-u64 coinbase incident and PASS on the fixed miner's canary (per-case batches, rule 26)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:02:23 +00:00
igneum-labs
2fc7d5c4d7 Merge box/master 36816182c into class-v6-floor-k-docs-2 (the steward's 36816182 under it; the repeat batch trimmed to floor lane 2's rows ADV-05 and ADV-06, method model; the registry master's with that batch recorded) 2026-10-08 21:02:08 +00:00
igneum-labs
11a06ccb4b Merge remote-tracking branch 'box/master' into ci-guest-format
# Conflicts:
#	docs/plans/igneum-2.0-test-registry.json
#	tools/ci/checks.txt
#	tools/ci/pre-push.sh
2026-10-08 21:02:04 +00:00
igneum-labs
a433836eb3 Merge master 36816182 into workers-page-nine under the master-landing lock 2026-10-08 21:01:32 +00:00
igneum-labs
85f6df197a Batches: the stale repeat shadow-k-20261008-rows-repeat removed (a stopped chain's commit; floor-k-20261008-rows-repeat is the one that stands, on floor lane 2's word) 2026-10-08 20:54:31 +00:00
igneum-labs
6a492df103 The recorder drops a lifted reason-only record when a real cell lands and never counts it in the combined decision (R2-F03-R01 read NOT RUN beside its PASS cell); the F03 batch re-recorded
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:53:54 +00:00
igneum-labs
6c5c3432cc Merge remote-tracking branch 'box/master' into ci-rule33-canary
# Conflicts:
#	docs/plans/igneum-2.0-test-harness-map.md
#	docs/plans/igneum-2.0-test-registry.json
#	tools/ci/batches/kills-20261008.json
#	tools/ci/test-map.json
2026-10-08 20:53:44 +00:00
igneum-labs
95c16e1eef Merge remote-tracking branch 'box/master' into ci-guest-format 2026-10-08 20:53:19 +00:00
igneum-labs
4b0d285781 /prize: the disclosure prize's terms (a placed or measured adversarial implementation beating the served bracket, or a reproduced shortcut; not confirmations, synthesis-only results, outside-budget work or excluded classes; judged in-house, the finding served either way); the phrase leaves the served forbidden list on the founder's word of 8 October; the audit's prize row and the three closed open items; the og prize sub line within 80
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:53:08 +00:00
igneum-labs
0f5d6eeca2 route-spill check: the unreachable-box probe points at port 1 (ssh URI), so a gate run on a build box reads it as down
On a box sshd answers on 127.0.0.1:22 and a forwarded agent logs in, so the bare address returned a slot line and the
gate was red on every box run while green on the Mac. The probe still exercises the same /bin/bash path through lib.sh.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:52:14 +00:00
igneum-labs
026563273d Merge master e2859af44 into ci-f03-manifest under the master-landing lock 2026-10-08 20:51:04 +00:00
igneum-labs
ce413d546d Merge commit 'e2859af44' into site-2.0-followup 2026-10-08 20:50:48 +00:00
igneum-labs
82777eff40 registry: OPS-03 manifest cell recorded with a real evidence path
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:50:48 +00:00
igneum-labs
d21e613b10 The guest input format moves with the pinned guests: a GUEST_INPUT_FORMAT bump without a change under proving/igneum-prove/elf/ is refused at the merge; the gate reads the elf manifest's guest_input_format against the source constant (a pre-provenance pin is a named line); self-tests
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:50:45 +00:00
igneum-labs
c2ab0c0a05 Merge master ebdb8916 into site-2.0-followup (registry: OPS-03 manifest row re-recorded)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:50:25 +00:00
igneum-labs
1d75fdf3c0 Merge master ebdb89163 into pool-registry under the master-landing lock 2026-10-08 20:48:39 +00:00
igneum-labs
8242739171 Test registry: the rows that name shadow-k.md moved by the repeat batch floor-k-20261008-rows-repeat (ADV-05, ADV-06 floor lane 2's on its word, method model; POW-03, POW-07 the steward's; no decision changed) 2026-10-08 20:46:23 +00:00
igneum-labs
c48cd4f885 Merge box/master ebdb89163 into class-v6-floor-k-docs-2 (rule 26: master's map, page and registry taken) 2026-10-08 20:46:22 +00:00
igneum-labs
764047f05d Rule 33: the fresh-install canary record (tools/ci/canary/<sha>.json, one file per sha with an artefacts list) and its check; publish-manifest.sh and publish-public.sh refuse an entry whose sha has no PASS record for its own artefact; the canary cell under INT-07 (BLOCKED tonight); the merge replays modified batches too (the kills re-record reaches master's rows); the identity words out of the map
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:46:09 +00:00
igneum-labs
9e65ae5023 Test registry: the rows that name shadow-k.md moved by the repeat batch shadow-k-20261008-rows-repeat at this landing (no decision changed; named per owner) 2026-10-08 20:45:54 +00:00
igneum-labs
e7b9ee3863 Merge box/master ebdb89163 into class-v6-floor-k-docs-2 (rule 26: master's map, page and registry taken) 2026-10-08 20:45:54 +00:00
igneum-labs
2f6e1dfb20 registry: R2-F03-R02 leaves the map's not-run table now that the kit's cell names it; recorded RUNNING
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:44:26 +00:00
igneum-labs
8ce08c5413 registry: the kit batch cites paths only in its evidence field (the page, zip 4, the emulation log on build-1) with the reading in a note; recorded RUNNING on R2-F03-R02
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:44:04 +00:00
igneum-labs
b8549cd60a REV and INT generators keep evidence_records, blocked_on and method_recorded on regeneration (a regenerated suite dropped every recorded cell; self-tests assert they survive)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:43:52 +00:00
igneum-labs
027c26b44b registry: the kit's cell names master's case id R2-F03-R02 (REV-F03-2 renamed on master) and the batch carries its method (GPU); re-recorded, the map page regenerated
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:43:47 +00:00
igneum-labs
d8c30f3800 Merge build/master 366452636 into class-v6-kits (master's registry, harness map and gate-remote.sh taken; the kit's batch kit-class-v6-20261008-01 re-recorded on master's registry and the map page regenerated)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:43:21 +00:00
igneum-labs
4812e95cb5 Workers page: one card per entry of the merged boxes[] (nine boxes), the Mac mini card, the main site icon set, no per-box feed map
The page drew build-1 to build-3 only: a hardcoded three-entry feed map fed the cards while merge-workers.mjs on build-1
already carried nine boxes. The map and its live-feed loader are gone; every card comes from the merged workers.json.
build-3 reads "down since 16:40 UK, Hetzner switch fault" when its feed is absent or stale; a box with no builds log yet
reads "provisioning, first build logged when the sources land". The MacBook card says nothing builds there except the one
DMG and the mini carries macOS binaries from tonight; igneum-mini (the Mac build box, M6) has a server row and a card that
read "no report yet" until its collector pushes. The head carries site/index.html icon links (the ember mark as an SVG
data URI, favicon.ico, favicon-32.png, icon-192/512, apple-touch-icon.png).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:43:14 +00:00
igneum-labs
e44410e7b9 F03 batch: GOV-01 cites the fork's freeze file as a box path (the recorder's tree rule)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:43:07 +00:00
igneum-labs
f71641130b chip-model: delete the simulation traces after each power report (a full pod disk killed four runs)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

Documents-only replay of 14d4f3a32 (86e5b0fb24) for the box mirror master
2026-10-08 20:42:29 +00:00
igneum-labs
816303c5b5 shadow-k: the placed gated 32-register base (6.7 pJ per lane-op ASAP7, the board 2.4x node-for-node and 2.8x a node ahead at the lock), the 8 KB scratch row (104 pJ per read at N3, k 0.074 against the L2 hit)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

Documents-only replay of 07029f7ee (86e5b0fb24) for the box mirror master
2026-10-08 20:42:29 +00:00
igneum-labs
13617b891c shadow-k: the adversary's gated rows (4.5 and 6.2 pJ per lane-op ASAP7; the window's residual +0.13 of k), the placed ungated base (11.3 pJ), the ICG simulation model
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

Documents-only replay of 9737ffb26 (86e5b0fb24) for the box mirror master
2026-10-08 20:42:28 +00:00
igneum-labs
0dc59c302e shadow-k: the window's measured GPU side (no spill, 83 and 67 percent occupancy, at most 5 percent per load); the time-multiplexed core, the connected-state draw, the simplified FP32 units and the pod mode
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

Documents-only replay of 26844fde7 (86e5b0fb24) for the box mirror master
2026-10-08 20:42:28 +00:00
igneum-labs
eb4bcb98e0 shadow-k: the adversary's 64-register core (the live-state analysis: 95 percent of the window necessary across every wait under the fold-reads-all rule; the state's cheaper forms modelled; the gated rows pending), the gated variants' configs, livestate.py
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

Documents-only replay of f1ddf0b39 (86e5b0fb24) for the box mirror master
2026-10-08 20:42:27 +00:00
igneum-labs
691914d751 Merge master 366452636 into ux-01-batch-2 under the master-landing lock 2026-10-08 20:42:22 +00:00
igneum-labs
6db5646a9c registry: pool batch 03 at 7ffd29af with its evidence page docs/analysis/pool/pool-pair-2026-10-08.md (UX-05 vote-key PASS in the crate; the devnet-4 live pair NOT RUN, its wrong_hash an unsynced node's unsettled class v5 state, not a generator skew; the guard on pool-2.0; two 2.0 gaps closed)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:41:26 +00:00
igneum-labs
26417c3ab6 Merge master fbade9172 into class-v6-adversary-docs-3 under the master-landing lock 2026-10-08 20:41:17 +00:00
igneum-labs
66ca5b4fa2 Merge master a08a017e0 into fleet-pidkill-20261008 under the master-landing lock 2026-10-08 20:40:03 +00:00
igneum-labs
b57801617c Merge master ba1e4b89d into class-v6-family-gate under the master-landing lock 2026-10-08 20:38:52 +00:00
igneum-labs
5cc5880be6 Merge remote-tracking branch 'build/master' into ux-01-batch-2 2026-10-08 20:35:25 +00:00
igneum-labs
5bf8e09707 Batch adversary-20261008-placed-8lane: adversary:d2b's evidence cited at the document (section 13); the mf results file stays on the branch until it lands (the recorder's tree rule since f7c6d511); the batch recorded on master's registry 2026-10-08 20:32:56 +00:00
igneum-labs
0c245a6083 fleet: every stop goes through a pid file; pkill and killall are gone from the live tree
The founder's word (20:21 BST 8 Oct 2026): pkill and killall refuse by construction on every box and pod (exit 97). lib/pidkill.sh
(kill_pidfile, kill_children by parent pid, kill_sock_owner through ss -xlp) is the shared form; box-prover.py finds the SP1 server by
the pid owning its socket; lib/box.py, box-kill.sh, kill-node.sh, the rig scripts (box-ember, box-rig, box-matrix, box-floor-v5),
lib/standing.py and publish-2-move.py stop by pid files; box-dn3.sh writes node, miner-loop and miner pids and uses pidkill;
fleet-puller.sh and dn3-kill.sh call fleet-stop.sh. Seventeen devnet-2-era scripts that only ever stopped by name move to
tools/fleet/retired/ with a README.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:32:25 +00:00
igneum-labs
acf161cd91 Merge box/master f7c6d5116 into class-v6-adversary-docs-3 (rule 26: master's map, page and registry taken; the adversary cells re-added with ADV-05; the batch recorded) 2026-10-08 20:32:17 +00:00
igneum-labs
41bce2693c Merge master f7c6d5116 into pool-guard-docs under the master-landing lock 2026-10-08 20:32:03 +00:00