Commit Graph
420 Commits
Author SHA1 Message Date
Kevin Veen-Birkenbach 2f25443a9f Merge branch 'feature/code-graph-by-default' into unify/skills-20261114
# Conflicts:
#	docs/requirements/009-code-graph-by-default.md
#	tests/test_install.py
2026-11-14 21:56:06 +00:00
Kevin Veen-Birkenbach ba67ebf9b1 Merge branch 'documentation/token-saving-requirements' into unify/skills-20261114 2026-11-14 21:53:38 +00:00
Kevin Veen-Birkenbach 50bfed6751 Merge branch 'fix/hub-speech-bubbles-hidden' into unify/skills-20261114
# Conflicts:
#	orchestrator/templates/page-bubbles.js
#	orchestrator/templates/page-speech.js
#	tests/test_hub_bubbles.py
2026-11-14 21:53:37 +00:00
Kevin Veen-Birkenbach cfdd83dd2e Merge branch 'fix/unify-e2e-bubble-bar-overlap' into unify/skills-20261114 2026-11-14 21:52:15 +00:00
Kevin Veen-Birkenbach 568747651b Merge branch 'fix/hub-cleanup-pkgmgr-local' into unify/skills-20261114 2026-11-14 21:52:08 +00:00
Kevin Veen-Birkenbach 6d1c28241b Merge branch 'fix/hub-operator-rows-close-stale' into unify/skills-20261114 2026-11-14 21:52:07 +00:00
Kevin Veen-Birkenbach 3af83b0c46 Merge branch 'fix/kvb-cleanup-rule-text' into unify/skills-20261114
# Conflicts:
#	skills/kvb-cleanup/SKILL.md
2026-11-14 21:52:07 +00:00
Kevin Veen-Birkenbach 064e540464 Merge branch 'fix/kvb-check-security-medusa-timeout' into unify/skills-20261114 2026-11-14 21:51:11 +00:00
Kevin Veen-Birkenbach 80ab621c53 Merge branch 'fix/kvb-org-operator-utf8-input' into unify/skills-20261114 2026-11-14 21:51:11 +00:00
Kevin Veen-Birkenbach 9d653aa9ce Merge branch 'fix/kvb-org-rasci-artifact-filters' into unify/skills-20261114 2026-11-14 21:51:10 +00:00
Kevin Veen-Birkenbach faded2ed54 Merge branch 'fix/kvb-git-hotfix-adopt-and-sync' into unify/skills-20261114 2026-11-14 21:51:10 +00:00
Kevin Veen-Birkenbach f315cf37e7 Merge branch 'fix/kvb-check-score-isolated-import' into unify/skills-20261114 2026-11-14 21:51:10 +00:00
Kevin Veen-Birkenbach 041851d296 Merge branch 'fix/kvb-org-queue-backfill' into unify/skills-20261114
# Conflicts:
#	skills/kvb-org-queue/SKILL.md
#	skills/kvb-org-queue/scripts/queue.py
2026-11-14 21:49:10 +00:00
Kevin Veen-Birkenbach 0046ab3a0a Merge branch 'fix/queue-pressure-gate' into unify/skills-20261114
# Conflicts:
#	scripts/install.sh
#	skills/kvb-org-organize/scripts/watch.py
2026-11-14 21:46:47 +00:00
Kevin Veen-Birkenbach ccb7373725 Merge branch 'fix/hook-exec-hardening' into unify/skills-20261114 2026-11-14 21:44:46 +00:00
Kevin Veen-Birkenbach cdd4df5234 Merge branch 'fix/council-root-durable' into unify/skills-20261114 2026-11-14 21:44:45 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 4ec1a43238 fix(orchestrator): approve only a plain commit of the exact staged index
The approval hook binds a hub approval to a plain git commit of the staged index only; commits that add files from the working tree, amend, take a pathspec, run in a chain or inside a shell wrapper fall back to the normal dialog. Every git call of the hook runs with an environment allowlist. Tests run once on unify/skills-20261114 after the merge (operator order 2026-11-14).

🎯 Fix: 93 %
💥 Break: 3 %
💸 Tokens: –
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 90
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 21:43:06 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 911eb521b9 feat(kvb-agent-optimize): initialise the code graph by default
graph.py indexes a repository into codebase-memory-mcp on first use and keeps auto_index on, so every session starts with a graph; agents.py reports its state. Tests run once on unify/skills-20261114 after the merge (operator order 2026-11-14).

🎯 Fix: 88 %
💥 Break: 5 %
💸 Tokens: –
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 84
🚦 Gate: ⛔ fix 88 % < 90

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 21:25:12 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 d882e501c1 fix(orchestrator): queue every speech row of a poll before one redraw
speakRows queues all speech rows of a poll before the single redraw, so a MAYDAY behind other rows is never lost, and a draw that throws shows a toast instead of stopping the pump. Late and stale rows both bubble. Tests run once on unify/skills-20261114 after the merge (operator order 2026-11-14).

🎯 Fix: 80 %
💥 Break: 5 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: 100 %
🏆 Score: 76
🚦 Gate: ⛔ fix 80 % < 90

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 21:23:12 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 23b6d8579c fix(orchestrator): mark a branch done only when pkgmgr deleted it
The cleanup runs pkgmgr branch cleanup --local and offers deletion only for ancestry-merged branches; squash or PR merges stay a dry-run preview. A signed branch whose ref vanished without a pkgmgr delete becomes gone (deleted outside this run, or after an interrupted apply) and is shown apart from deleted and failed; an Unknown --branch answer for a vanished ref repeats the run once without it. Tests run once on unify/skills-20261114 after the merge (operator order 2026-11-14).

🎯 Fix: 88 %
💥 Break: 6 %
💸 Tokens: –
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 83
🚦 Gate: ⛔ fix 88 % < 90, break 6 % > 5

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 21:22:28 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 ec30d8c94e fix(orchestrator): close operator rows the server no longer has
A 404 on Run, Close, a passkey signature or a rule accept closes the card with the notice "already done or removed elsewhere" instead of a lasting error; a selection reloads /api/state once and closes only the cards that are gone, other failures keep the old error. Tests run once on unify/skills-20261114 after the merge (operator order 2026-11-14).

🎯 Fix: 90 %
💥 Break: 3 %
💸 Tokens: –
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 87
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 21:21:07 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 8726ed37a7 fix(kvb-org-organize): move the council root out of /tmp with a lossless migration
councilroot.py resolves one durable council root for every script and migrates the old /tmp root under the held locks: a preflight conflict moves nothing, each legacy file is renamed to a staging name before it is copied so a late writer is merged by the catch-up, id-less journals merge by line multiset, a differing whole file is kept beside its namesake, and a crash marker makes install copy the skills and report. One settings-file list, managed drop-ins included, serves councilroot.py and queue.py. Tests run once on unify/skills-20261114 after the merge (operator order 2026-11-14).

🎯 Fix: 88 %
💥 Break: 7 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: 100 %
🏆 Score: 82
🚦 Gate: ⛔ fix 88 % < 90, break 7 % > 5

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 21:18:41 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 9fc1383e97 feat(kvb-org-queue): let jobs pass holders that can never start and help a disk-blocked head
A waiting job passes a same-line holder that cannot start under the current caps and takes its place; the decision is recomputed every pass, so a holder that recovers regains its place. A hotfix holds others back and heads the line only while it could start. A holder blocked by disk alone gets the cleanup and the Main notice even while smaller jobs backfill. Main-first reservations stay; the replay pins the lines bounds and caps its mean. Tests run once on unify/skills-20261114 after the merge (operator order 2026-11-14).

🎯 Fix: 80 %
💥 Break: 12 %
💸 Tokens: 100 %
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 70
🚦 Gate: ⛔ fix 80 % < 90, break 12 % > 5

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 21:17:30 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 8e2db852e5 feat(kvb-org-organize): hold heavy jobs back on memory pressure with one shipped default config
watch.py admits heavy work only while PSI, swap-out rate and RAM forecast stay under the configured limits; config.default.json is the single default the install writes once (mode 600, no clobber) and every hook, admit and queue command falls back to. Browser runners and package-manager script forms count as heavy, preflight derives its probes from the light and heavy tool lists, and the shipped default carries no prune command, buildx path or host-specific key. Tests run once on unify/skills-20261114 after the merge (operator order 2026-11-14).

🎯 Fix: 85 %
💥 Break: 8 %
💸 Tokens: 99 %
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 79
🚦 Gate: ⛔ fix 85 % < 90, break 8 % > 5

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 21:17:28 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 699f3227c9 fix(tests): keep the holder escape test free of non-English characters
The non-ASCII printable case of the holder escape test wrote a literal umlaut into tests/test_gate.py, which the repository-wide English check rejects. It now uses an escaped character, so the case still proves that printable non-ASCII text reaches the message unchanged.

🎯 Fix: 98 %
💥 Break: 1 %
💸 Tokens: 100 %
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 97
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 19:52:41 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 3632086ea0 fix(kvb-check-gate): escape control characters in the recorded lock holder
A busy message printed the holder line of a lock file raw, so a recorded command carrying terminal escapes or newlines could forge stderr lines or move the cursor in another agent's terminal. holder() now shows every non-printable character as its escape sequence and names no holder when the recorded pid is not a number.

🎯 Fix: 95 %
💥 Break: 2 %
💸 Tokens: 100 %
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 93
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 19:25:49 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 bdb3947ec7 fix(kvb-cleanup): write down the seven cleanup rules runs kept tripping on
Cleanup runs kept stumbling on the same seven points: recent items an owner had already released, fuser exit codes, empty Other answers, find -exec rm that the classifier refuses, merged branches whose tasks count as stale, bfs failing on -newermt, and one-option questions that AskUserQuestion rejects. SKILL.md and execute.md now state each rule where the step needs it, within the 600-word cap and at 99 % of the previous token cost.

🎯 Fix: 90 %
💥 Break: 4 %
💸 Tokens: 99 %
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 87
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 18:41:48 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 51c04da936 fix(kvb-check-gate): hold gate locks that a symlink or a read-only git dir cannot break
gate.py took its run lock inside the git common dir only, so a linked worktree on a read-only git dir could not gate, and a planted symlink or hard link could steer the lock file. Each run now holds a per-worktree lock under /tmp/agents-locks/<sha16 of the toplevel> and a permanent kvb-gate-run.lock in the git dir, created 0600 with a read-only fallback. Every lock opens through an O_NOFOLLOW openat chain and is checked with fstat for owner and a single link. The state lock fails before any target runs, and hold passes only the two lock descriptors to its child. hotfix.py takes the same lock for its sync.

🎯 Fix: 93 %
💥 Break: 4 %
💸 Tokens: 100 %
⏱️ Velocity: –
🛡️ Security: 100 %
🏆 Score: 89
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 18:40:46 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 f145a600fe fix(kvb-check-security): give medusa a configurable scanner timeout and fail closed
medusa-security caps every scanner at a fixed 60 s per file, so long files
lost scanner results while the run still scored 100 %. security.py now
passes --timeout (MEDUSA_SCANNER_TIMEOUT) and an overall --deadline to the
medusa run (1 to 86400 s), runs medusa in its own process group that dies
with security.py (stop signals, exceptions, PR_SET_PDEATHSIG), keeps
signals the caller ignores ignored, and ends an incomplete or killed scan
with exit 4, nothing cached and nothing scored. The fault text names only
causes medusa can tell apart, and the cache key holds the timeout medusa
actually got.

🎯 Fix: 92 %
💥 Break: 4 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: 100 %
🏆 Score: 88
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 17:45:17 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 2ed3628d9b fix(kvb-org-operator): keep UTF-8 input whole at every prompt
operator.py run crashed with UnicodeDecodeError when the operator erased
an umlaut on a terminal without IUTF8: the erase key removed only the
second byte of the character and left a lone 0xc3 for input() to decode
strictly. typed() and timed() now set IUTF8 on a terminal stdin while
they wait for a line and restore the previous mode afterwards, and run
replaces undecodable bytes instead of crashing.

🎯 Fix: 95 %
💥 Break: 2 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: ~100 %
🏆 Score: 93
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 17:31:05 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 fec77026ea docs(requirements): add 009 code graph, 010 security cache, 011 test cache
Token-saving requirements picked by the operator on 2026-11-14: the code
graph is initialised automatically per repository, security scan results
are cached per file by content, and green test results are reused only
under an exact key. Implementation follows once the running lanes land;
the shared result store is requirement 003 of the quality repository.

🎯 Fix: 95 %
💥 Break: 1 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: 100 %
🏆 Score: 94
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 14:10:53 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 4cef28ee19 fix(orchestrator): pace speech bubbles by voice and reading, never backwards
voicePace and readingPace set each row's floor, span and duration in one place: the spoken rest reveals the unread words over the speech audio, a procedure word reveals none, a burst and silent reading reveal the rest at reading pace, and wordsRead never goes below the floor. A sound that fails or is dropped still ends cleared. Requirement 003 W6 states it; probes cover quiet, burst, fallback, cut, auto, check, silent, rejected and faulty rows.

🎯 Fix: 92 %
💥 Break: 4 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: 100 %
🏆 Score: 88
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 09:58:46 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 c297ad0890 fix(kvb-org-rasci): filter artifacts by worker, refuse a kindless drop
rasci.py artifacts takes --worker <text> to list only that worker's items. set refuses an --artifact or --drop-artifact without <kind>:<ref> (it used to no-op silently on a bare path), and a well-formed drop that is not recorded drops nothing and says so on stderr, so idempotent teardown runs keep working.

🎯 Fix: 93 %
💥 Break: 3 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: 100 %
🏆 Score: 90
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 09:57:16 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 d9ccb74e3d fix(kvb-git-hotfix): adopt the register task the hotfix belongs to
hotfix.py add now adopts an existing rasci task of the same branch and repository: the adding session becomes R and A, the former A keeps its other letters, and a task on another branch or repository is refused instead of overwritten.

🎯 Fix: 92 %
💥 Break: 4 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: 100 %
🏆 Score: 88
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 09:55:51 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 5544f83679 fix(kvb-check-score): import sibling modules under python -I
score.py and backfill.py ran with python -I drop their own directory from sys.path, so the sibling imports failed (Security 2.0 hit it 2026-11-14 02:40). Both now put their resolved script directory first; tests run both scripts isolated.

🎯 Fix: 95 %
💥 Break: 2 %
💸 Tokens: 100 %
⏱️ Velocity: 100 %
🛡️ Security: 100 %
🏆 Score: 93
🚦 Gate: ✅ 90/5, tokens and velocity <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 09:54:01 +00:00
Kevin Veen-Birkenbach a2a03995cb Merge branch 'fix/unify-e2e-firefox-artifact-popup' into unify/skills-20261112-night
🧪 Test / 🧪 Lock + lint (push) Canceled after 0s
2026-11-14 00:39:55 +00:00
Kevin Veen-Birkenbach 246716ece0 Merge branch 'fix/unify-e2e-firefox-artifact-popup' into unify/skills-20261112-night 2026-11-13 23:23:57 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 42102cf9d2 test(e2e): click outside the popup where the dialog takes the hit
The popup probe clicked elementFromPoint(2, innerHeight - 2) to close the popup. In Firefox at 1920 px that corner returned the html element in 4 of 10 stress rounds, so the close handler, which acts only on a click on the dialog itself, correctly ignored it. The probe now clicks the middle of the widest free margin around the popup box and waits up to 3 s until that point hits the dialog, else clicks whatever is there, so a real overlay still fails.

🎯 Fix: 85 %
💥 Break: 5 %
💸 Cost: –
⏱️ Time: –
🛡️ Security: 100 %
🏆 Score: 81
🚦 Gate: ⛔ fix 85 % < 90 (operator word 2026-11-13: commit and merge)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 23:23:51 +00:00
Kevin Veen-Birkenbach f18be035a4 Merge branch 'fix/unify-e2e-home-star-clip' into unify/skills-20261112-night 2026-11-13 23:19:08 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 459d8a82e0 test(e2e): count only the visible part of a star and a control as cover
At 390 px with the radio spot open on Agents Orbit, the STARS probe reported the legend star covered by the radio search field and its clear button. Both sit scrolled out of view inside the radio spot's overflow box, and a hit test at their points returns the legend, so nothing covers the star. The probe now intersects each box with every clipping ancestor before it measures overlap; it can only shrink a box, so a visible overlap is still reported.

🎯 Fix: 93 %
💥 Break: 4 %
💸 Cost: –
⏱️ Time: –
🛡️ Security: 100 %
🏆 Score: 89
🚦 Gate: ✅ 90/5, cost and time <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 23:18:07 +00:00
Kevin Veen-Birkenbach c202429cf9 Merge branch 'fix/unify-e2e-orchestrator-radio' into unify/skills-20261112-night 2026-11-13 21:35:36 +00:00
Kevin Veen-Birkenbach 74d9feed87 Merge branch 'fix/unify-e2e-home-tab' into unify/skills-20261112-night 2026-11-13 21:35:35 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 4c3219549f fix(orchestrator): keep 008's star from squeezing the graph cards at 390
The 008 star row took about 26 px from the stage cards: the tasks/3d canvas fell under 80 px, where zoomToFit's 40 px padding gave a negative field of view and froze the camera, and the agents/orbit card pushed the page past 844 px with the radio spot open, against 001's no-scrollbar rule. The chart legend is now the panel's star slot (style-home.css places it at the end of the last key line under 640 px), and the 3D fit clamps its padding to a quarter of the canvas height and skips a canvas of height 0. The Firefox radio probe waits for paused with src cleared, since Firefox pauses the element before the product's 3 s fade ends.

🎯 Fix: 92 %
💥 Break: 4 %
💸 Cost: –
⏱️ Time: –
🛡️ Security: 100 %
🏆 Score: 88
🚦 Gate: ✅ 90/5, cost and time <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 21:35:29 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 702acfb754 test(e2e): follow 008's Home tab in the walk, gaps, home and W4 tests
Requirement 008 put Home first in nav.js, so SCREENS and the walk badges gain home, key 1 now opens #home and Resources moves to key 6 (page-keyboard.js numbers keys by tab order). TestHomeDashboard borrowed TestOrchestratorE2E.setUp, whose zero-argument super() raised TypeError, so it never ran; the fixture is now TestOrchestratorE2E.seed(), called by both. Its first real run needed the two-argument speech stub, the gone-tile step at the last width, an overlap check that skips closed details menus and popup layers, and the grid expectation style-home.css really computes.

🎯 Fix: 93 %
💥 Break: 3 %
💸 Cost: –
⏱️ Time: –
🛡️ Security: 100 %
🏆 Score: 90
🚦 Gate: ✅ 90/5, cost and time <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 21:34:41 +00:00
Kevin Veen-Birkenbach 1356e2690e Merge branch 'fix/unify-e2e-firefox-csp' into unify/skills-20261112-night 2026-11-13 20:41:24 +00:00
Kevin Veen-Birkenbach 86b84bfb4e Merge branch 'fix/unify-e2e-speech-bubbles' into unify/skills-20261112-night 2026-11-13 20:41:23 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 908bd094e0 fix(orchestrator): drop the 3D library's nonce-less injected styles
The vendored 3d-force-graph appends three empty style elements to head while it loads and fills them with its own CSS; the page CSP allows styles by nonce only, so it blocks them and Firefox logs six style-src-elem errors, which failed the W3 artifacts e2e once 594d185 opened the 3D view there. page-graph3d.js now intercepts head.appendChild only while that script runs (document.currentScript is its tag), skips style elements from it and removes the interception on load or error. The CSP and the pinned library stay unchanged; the page already styles the classes those blocks targeted.

🎯 Fix: 95 %
💥 Break: 3 %
💸 Cost: –
⏱️ Time: –
🛡️ Security: 100 %
🏆 Score: 92
🚦 Gate: ✅ 90/5, cost and time <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 20:41:18 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 1f8f17b182 test(e2e): follow 004's browser-side speech settings in the bubble tests
Requirement 004 (acf3925) moved speech on, classes and muted into each browser's localStorage and gave the speech tool a second argument admit; the bubble-faces fixture still saved the old hub settings and stubbed the one-argument tool, which settings.py now refuses. The fixture saves only bubbles, stubs (body, admit) and sets the radio to Silence with mayday and security through a start-up script. W6's long row moves from security to broadcast, the class 004 switched W6 to, so the browser picks it again.

🎯 Fix: 95 %
💥 Break: 2 %
💸 Cost: –
⏱️ Time: –
🛡️ Security: 100 %
🏆 Score: 93
🚦 Gate: ✅ 90/5, cost and time <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 20:40:57 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 cd0bb3ccd0 test(orchestrator): give each fake radio.py its own mtime second
The radio_module test wrote five fake radio.py versions with nanosecond mtimes, all in second 0. Versions 5 and 6 have the same size, so with bytecode writing on (make test) CPython's pyc check reused version 5 for version 6 and the test failed only in the full suite. The installed hub is not affected: install.sh drops every pyc, the container sets PYTHONDONTWRITEBYTECODE and mounts /skills read-only.

🎯 Fix: 95 %
💥 Break: 1 %
💸 Cost: –
⏱️ Time: –
🛡️ Security: 100 %
🏆 Score: 94
🚦 Gate: ✅ 90/5, cost and time <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 20:40:26 +00:00
Kevin Veen-BirkenbachandClaude Opus 5.5 1e5aac8a37 fix(unify): repair what the night merges broke between lanes
937c2c0 added the line prefix KEPT to watch.py, which shadowed the state directory path watch.KEPT that gates.py and the queue cleanup use; the prefix is now KEPT_LINE. The receipts page test gets an identity revealed() since page-dom.js wraps names through it after the firewall merge, the operator tasks test expects the Home tab of 008, and the hub view id fixture uses a non-ASCII word the English check accepts. Accepted MEDUSA false positives, re-keyed by the rename: MM-SCAN-058 at watch.py:1417 and WM-SCAN-354 at watch.py:1418.

🎯 Fix: 92 %
💥 Break: 2 %
💸 Cost: –
⏱️ Time: –
🛡️ Security: 100 %
🏆 Score: 90
🚦 Gate: ✅ 90/5, cost and time <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:55:22 +00:00