Commit Graph

2173 Commits

Author SHA1 Message Date
Vincent Koc
b45ca07ec9 fix(ci): keep full release validation aligned with runtime state (#116653)
Align Telegram artifact reuse and packaged Docker validation with current GitHub Actions and runtime state contracts. Refresh affected regression fixtures and isolate test state.
2026-07-31 23:03:02 +08:00
Vincent Koc
b7eceb4595 fix(ci): target Testbox VM SSH sockets 2026-07-31 22:14:56 +08:00
Vincent Koc
a133f63488 fix(ci): preserve Testbox cancellation finalization 2026-07-31 22:14:56 +08:00
Vincent Koc
b702dcf4c5 fix(ci): close inbound Testbox SSH sessions 2026-07-31 22:14:56 +08:00
Vincent Koc
e47a1c3e1c fix(ci): close Testbox client sockets 2026-07-31 22:14:56 +08:00
Vincent Koc
3d2dbec9ce fix(ci): close Testbox SSH sessions on teardown 2026-07-31 22:14:56 +08:00
Vincent Koc
bd997c686f fix(ci): reuse Testbox dependency snapshot 2026-07-31 22:14:56 +08:00
Vincent Koc
33fd59a4bb fix(release): skip unavailable Buzz QA runner (#116753) 2026-07-31 18:51:00 +08:00
Vincent Koc
44ce19821f fix(ci): retire Kova runtime deps coverage (#116760) 2026-07-31 18:09:17 +08:00
Vincent Koc
37b1a71654 fix(release): unblock beta validation workflows (#116708)
* fix(release): repair beta validation workflow state

* fix(qa): dispatch yielded fanout runs without task tracking

* fix(test): honor GPT-Live shard opt-in
2026-07-31 15:47:11 +08:00
Dallin Romney
640b4ac599 fix(ci): discover frozen target startup benchmark cases (#116711) 2026-07-31 15:35:33 +08:00
Vincent Koc
c3bb8fc265 fix(perf): make Kova startup diagnostics actionable (#116319)
* fix(perf): rank startup duration hotspots correctly

* ci(perf): exercise prepared runtime startup cases
2026-07-31 10:23:32 +08:00
Patrick Erichsen
61cf4b9246 test(buzz): add live QA Lab channel lane (#116298)
* feat(buzz): add live QA runner

* ci(qa): add gated Buzz live lane

* docs(qa): document Buzz live testing

* fix(buzz): keep QA relay type private

* test(buzz): type QA suite mock

* fix(buzz): preserve racing QA reply relations

* fix(qa): harden Buzz credential redaction

* fix(buzz): ignore retained QA relay events

* fix(buzz): clean up observer setup failures

* test(buzz): return complete racing messages

* fix(buzz): default QA credentials to file

* fix(qa): validate pooled Buzz secrets

* fix(qa): redact structured Buzz auth tags

* fix(buzz): preserve QA credential env override

* fix(qa): redact inspected Buzz auth tags

* test(buzz): limit live QA to supported scenarios

* ci(qa): run Buzz lane in release checks

* fix(qa): require secure remote Buzz relays

* fix(qa): satisfy Buzz QA validation gates

* fix(qa): inject credential leasing into transport adapters
2026-07-30 17:44:27 -07:00
Vincent Koc
a473fc6f3e fix(ci): extend OpenAI live profile setup budget (#116580) 2026-07-30 23:40:02 +00:00
Dallin Romney
c06191e6fb fix(ci): pass OpenAI credentials to Matrix live frontier (#116434) 2026-07-30 23:40:06 +08:00
Vincent Koc
90a22b4f50 chore(test): run browser copilot E2E in CI (#116407)
* test(browser): run copilot E2E in CI

* test(browser): stabilize copilot E2E synchronization

* test(browser): wake extension workers deterministically

* test(ci): route browser copilot command changes
2026-07-30 23:20:34 +08:00
Dallin Romney
c854a04696 fix(ci): validate frozen release candidates faithfully (#116357) 2026-07-30 18:57:32 +08:00
Dallin Romney
bf8da2537c improve(ci): fail_fast option for release validation, off by default (#116352)
* ci: make release validation fail-fast optional

* ci: pass release QA fail-fast as boolean

* test: format release validation helper test
2026-07-30 18:53:41 +08:00
Dallin Romney
4edc0e9d9e fix(ci): pass target SHA to Telegram evidence finalizer (#116331) 2026-07-30 17:31:45 +08:00
Dallin Romney
2e26244cec fix(ci): support legacy Telegram release candidates (#116305)
* fix(ci): support legacy Telegram release candidates

* fix(ci): bound Telegram legacy evidence exception
2026-07-30 16:52:31 +08:00
joshavant
28f73d93c1 fix(ios): isolate capability router handlers 2026-07-30 03:04:42 -05:00
Dallin Romney
c8f0db2c24 fix(ci): load frozen shard runner without repository fetch (#116273)
* fix(ci): load frozen shard runner without repository fetch

* test(ci): preserve workflow expression delimiters

* fix(ci): use non-cone frozen runner checkout
2026-07-30 15:37:44 +08:00
Peter Steinberger
2c436594de fix: clean npm preflight cache restore targets (#116276) 2026-07-30 00:33:55 -07:00
Dallin Romney
6e60f48bad fix(ci): accept one verified Telegram SUT launch (#116267) 2026-07-30 14:42:46 +08:00
Dallin Romney
47a21192fd fix(ci): let trusted Telegram QA stage runtime (#116252) 2026-07-30 13:45:01 +08:00
Peter Steinberger
5a3c3b3f4f fix(ci): explicitly serialize hosted Swift tests (#116167) 2026-07-29 18:55:37 -07:00
Vincent Koc
1065c1d697 fix(ui): identify source builds in settings footer (#116134)
* fix(ui): identify source builds in settings footer

* fix(release): classify all official Control UI builds
2026-07-30 08:58:34 +08:00
Peter Steinberger
1165be39ff ci: serialize hosted macOS Swift tests (#116137) 2026-07-29 17:40:41 -07:00
Peter Steinberger
e8524a5839 feat: add Fish Audio S2.1 and local MLX speech (#115790)
* feat: add Fish Audio speech support

* chore: remove unused speech exports

* chore: keep Fish API constant private

* test: remove stale code mode import

* style: use bracket access for Fish voice ids

* test: restore QA evidence validator import
2026-07-29 05:15:33 -04:00
Dallin Romney
0723c68b57 refactor(qa): remove execution sharding (#115787) 2026-07-29 17:06:57 +08:00
Dallin Romney
f027d7bcf6 refactor(qa): make Matrix sharding execution-owned (#115589)
* refactor(qa): make sharding execution-owned

* test(qa): simplify transport selection coverage
2026-07-29 15:23:37 +08:00
Dallin Romney
28e09a04d3 ci: allow QA evidence generation to run for 90 minutes 2026-07-29 15:22:16 +08:00
Dallin Romney
7fe920a7d0 fix(ci): grant Telegram QA runtime scratch access (#115659) 2026-07-29 14:42:43 +08:00
Peter Steinberger
117e44cb80 fix(linux): package media codecs for deb and AppImage (#115616)
* fix(linux): package media codecs

* fix(linux): bundle AppImage media codecs

* docs: refresh Linux page map
2026-07-29 02:03:55 -04:00
Dallin Romney
2fa91f48d0 fix(ci): validate frozen release candidates (#115605)
* fix(ci): support historical i18n targets

* test(ci): cover historical i18n verification
2026-07-29 13:43:39 +08:00
Dallin Romney
aca7b59526 refactor: make Telegram release workflow tests behavioral (#114023)
* refactor: make Telegram release workflow tests behavioral

* test: fix Telegram release workflow CI
2026-07-29 13:23:35 +08:00
mikasa
887b37cf80 fix(ci): bound Mantis Discord fetches (#110851)
Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-07-28 23:36:27 -04:00
Dallin Romney
0d8ee3a1c5 fix(qa): validate profile evidence against scenario inventory (#115446) 2026-07-29 11:29:56 +08:00
Peter Steinberger
a8fb4872fb refactor: harden shell completion and shard Control UI E2E (#115332)
* refactor: harden CLI completion and shard Control UI E2E

* fix: keep completion command-tree result private

* fix: remove unused browser test helper exports

* test: cover sharded Control UI workflow contract
2026-07-28 14:35:27 -04:00
Peter Steinberger
4683c752ed test(ui): stabilize browser route readiness (#115320) 2026-07-28 13:59:05 -04:00
Peter Steinberger
1a13bc5c54 fix(ci): validate frozen Matrix QA targets (#115298)
* fix(ci): support frozen Matrix QA profiles

* style(ci): format Matrix workflow contract test
2026-07-28 13:22:56 -04:00
Yue Fei
1b4a465ea1 feat(gateway): accept WebSocket request trace context (#113189)
* feat(gateway): accept WebSocket request trace context

* test(gateway): prove WebSocket trace isolation

* test(gateway): simplify traced response match

* ci: allow iOS screenshot validation to finish

* fix(gateway): keep traced request failures scoped

* test(ci): align iOS screenshot timeout contract

* test(ui): reset config route location

* ci: scope iOS screenshots to native changes

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-07-28 13:15:05 -04:00
Vincent Koc
8f40fac955 fix(ci): run localization checks for UI changes 2026-07-28 23:47:14 +08:00
Peter Steinberger
3224005c82 fix(ci): bound Telegram Mantis proof cleanup (#115253)
* test(mantis): bound Telegram proof cleanup

* test(mantis): enforce cleanup deadline in wrapper

* test(mantis): supervise cleanup through bash
2026-07-28 11:44:28 -04:00
Dallin Romney
9feb6ad161 refactor(qa): make scenario selection taxonomy-owned (#114055)
* refactor(qa): remove scenario execution profiles

* fix(qa): preserve smoke CI membership

* test(qa): drop redundant smoke pack order check

* refactor(qa): remove Matrix runner profiles

* refactor(qa): simplify Matrix CI orchestration

* refactor(qa): derive execution sets from taxonomy

* fix(qa): keep Matrix shard helper private

* refactor(qa): derive profile scenarios from taxonomy

* refactor(qa): reconcile taxonomy selection

* refactor(qa): use canonical profile planning

* refactor(qa): select canonical coverage owners

* fix(qa): keep profile selector internal

* refactor(qa): remove representative coverage owners
2026-07-28 22:23:41 +08:00
Dallin Romney
a35bf77c82 fix(release): run harness install from its cwd (#115172) 2026-07-28 20:20:32 +08:00
ToToKr
d57e228c3a docs(mistral): fix broken adjustable reasoning docs URL (#114867)
* docs(mistral): fix broken adjustable reasoning docs URL

The linked Mistral page 404s; the reasoning docs now live one path segment up. Verified: old URL returns HTTP 404, new URL returns HTTP 200 and documents reasoning_effort. oxfmt --check passed on the changed file.

* fix(ci): audit links inside docs components

* test(cli): isolate root help exit fixture

* test(ci): split embedded agent shard

* fix(ci): harden docs link projection

* test(ci): satisfy docs audit strict types

* test(ci): isolate incomplete-turn harness

* test(cli): document root help fixture boundary

* test(ci): isolate overflow compaction harness

* test(ci): isolate contaminated agent model tests

* test(ci): update isolated prefix routing

* test(ci): bound loaded streaming HTTP cases

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-07-28 06:55:02 -04:00
Vincent Koc
ac157b3af0 fix(openshell): restore sandbox creation on current OpenShell (#115058)
* test(openshell): use structured gateway discovery

* fix(openshell): bound sandbox runtime names

* test(openshell): bound policy sandbox fixture

* docs(sandbox): document compatibility matrix

* fix(openshell): preserve registered runtime identities

* fix(openshell): quote recreate scope safely

* fix(openshell): match shipped legacy names

* docs(sandbox): refresh generated map

* test(sandbox): mock registered runtime lookup

* fix(openshell): reject non-ready legacy runtimes
2026-07-28 11:13:04 +02:00
Peter Steinberger
c141496217 fix: full release validation avoids embedded-agent shard stalls (#115019)
* ci: use scoped embedded-agent timeouts

* ci: align frozen-target test watchdogs

* ci: align embedded-agent timeout assertions
2026-07-28 04:38:32 -04:00
Peter Steinberger
1e3a4a69b5 fix: preserve native trace heading typography (#103436)
* fix(ios): preserve native trace heading typography

* test(ios): attach trace heading visual proof

* test(ios): import shared chat UI for visual proof
2026-07-28 00:01:44 -04:00