* feat(android): add Wear proxy protocol
Introduce a bounded, versioned phone/watch RPC contract and wire its test, lint, and build gates before either runtime endpoint lands.
Co-authored-by: Sebastian Schubotz <git@sibbl.de>
* test(android): lock Wear wire names
Cover every supported RPC method and event discriminator so phone and watch cannot silently drift.
Co-authored-by: Sebastian Schubotz <git@sibbl.net>
* fix(android): bound Wear JSON depth
Reject excessively nested Data Layer envelopes before kotlinx.serialization can recurse through them.
* fix(android): enforce Wear depth on encode
Keep outbound messages inside the same JSON nesting contract enforced by the decoder.
* fix(android): preflight Wear payload depth
Traverse arbitrary JSON payloads iteratively before serialization so deeply nested local trees fail safely.
---------
Co-authored-by: Sebastian Schubotz <git@sibbl.de>
Co-authored-by: Sebastian Schubotz <git@sibbl.net>
* perf(ci): carry extension boundary artifacts on a Blacksmith sticky disk
The boundary lane spends ~113s of its 214s rebuilding plugin-sdk
declaration artifacts that an actions/cache was meant to carry — but
that cache is evicted so fast under the repo quota that even its prefix
restore never hits. Same-repo runs now keep the artifacts on a sticky
disk with a coarse toolchain key (source changes intentionally hit a
stale snapshot: restored mtimes predate the fresh checkout, so the
mtime-based freshness predicate always routes into the incremental
rebuild — a stale snapshot can never false-skip the compile). Fork PRs
keep the GitHub cache path and can never write repository-global
snapshots. Seed/restore round-trip verified locally with rsync -aR
layout mirroring.
* perf(ci): restore boundary artifacts for the lint lane too
check-lint's shard runner spawns the same plugin-sdk boundary artifact
prep the boundary lane snapshots (~72s of its wall clock); it now
mounts the shared sticky read-only-by-convention (only the boundary
lane seeds) and restores before the run step.
* fix(ci): make the lint boundary-sticky mount strictly read-only
Autoreview caught a first-hydration race: with commit if-missing, a
cold-key lint job racing the boundary lane could commit an empty,
marker-less snapshot that if-missing then never repairs. commit false
makes lint a pure reader; only the boundary lane can seed.
* perf(ci): unblock parallel oxlint shards and gate the lint-lane i18n verify
check-lint is the slowest static lane (median 252s): the control-UI
i18n verify runs 86s unconditionally and the three oxlint shards run
serially (89s) because the parallel gate demands 48GB of memory — a
local-Mac threshold that Blacksmith's 32GB 16 vCPU class can never
meet. CI now parallelizes shards at 8+ cores / 24GB+, and the lint case
reuses the existing run_control_ui_i18n manifest gate so diffs that
cannot affect generated catalogs skip the verify while oxlint always
runs.
* chore(ci): print oxlint shard concurrency decision
The lint-lane validation run still executed shards serially on a 16
vCPU runner; one log line exposes the resolved concurrency and the
host resources the decision saw.
check-prompt-snapshots costs 200-310s per PR regenerating real prompt
stacks, but snapshots only change when the generator's import graph or
its fixtures do. The manifest now computes that reach (same pattern as
QA smoke gating: always-run surface regex incl. the codex extension
whose test API loads via a dynamic module id, plus import-graph walk
from the snapshot helper; deletions and null diffs fail safe to
running). Unaffected PR diffs skip the lane with an explicit log line;
pushes and dispatches always run.
* perf(ci): scale Vitest workers with runner class and share serial module cache
Compact node-test jobs pinned OPENCLAW_VITEST_MAX_WORKERS=2 regardless of
runner size, so import-bound suites (30-45s of module-graph import per
file) crawled: runner-cli spent 284s on 5s of tests, commands-1 296s on
10s. With serial plans the budget now scales with the class (6 on 8vcpu,
3 on 4vcpu); timing-sensitive groups (tooling, tui-pty, infra-process)
stay pinned to 2 via plan-level env. Serial bins share one Vitest
fs-module-cache path so later plans reuse the first plan's transforms.
The import-bound commands stripes and security suite move to the 8vcpu
class, and packing hints refresh from serial-run measurements (run
29481835688).
* perf(ci): stripe cli-runner suite and pin media-ui worker budget
The agents-core config runs files serially (fileParallelism false guards
shared module state), so raising the worker budget cannot help its
import-heavy cli-runner suite (~35s module import per file, 213s serial).
Stripe it three ways so bins parallelize the imports instead. media-ui
hosts browser locator tests that timed out at 6 workers on the first
validation run; pin it to the proven 2-worker budget. commands-1's hint
drops to its measured 6-worker runtime.
* perf(ci): diff-gate QA smoke by CLI import graph and isolate prompt snapshots lane
* fix(ci): run prompt snapshot lane unconditionally
* fix(ci): treat QA lane orchestration files as QA-impacting
The Crestodian->OpenClaw reconciliation restored a pre-#102197 `if (opts.approved)`
wrapper around the execute-time plugin-install trust check, so an unapproved
plugin-install op produced a formatted plan for an arbitrary npm/url/file
source instead of rejecting it. Validate unconditionally, matching #102197,
and restore its dropped regression test. Also fix a stale
crestodian-ring-zero-setup id in the CI compat list and a doubled
openclaw-openclaw-planner tmpdir prefix from the mechanical rename.
* ci: enforce changed-file TypeScript LOC ratchet
* ci: derive release-gate LOC base from PR
* test(ci): exclude repository test helpers from LOC ratchet
* ci: validate LOC ratchet on PR merge tree
* style: format release maintainer skill
* ci: run LOC ratchet for fast-only changes
* fix(ci): harden LOC ratchet comparisons
* fix(ci): cover native TypeScript in LOC ratchet
* fix(ci): compare LOC against tested merge tree
* test(ci): cover LOC manifest routing
Remove the hard 500-line TypeScript LOC ratchet from CI and local check planning, along with its package commands, implementation, baseline, and dedicated tests. Keep the remaining pull-request temp-creation report on its own pr-base fetch.\n\nVerification: 99 focused tooling tests; actionlint; zizmor; workflow guards; exact-head autoreview.
* chore(deadcode): add generated knip unused-export baseline
* ci(deadcode): replace broken export report lanes with enforced knip exports ratchet
* chore(deadcode): regenerate unused-export baseline on current main
* chore(deadcode): regenerate unused-export baseline on current main
* fix(ui): make indexed access explicit across the Control UI
Burns down all 322 ui-lane noUncheckedIndexedAccess errors: untrusted
markdown/diff/patch parsing gets miss-tolerant guards (renderer rules
degrade to empty output instead of throwing mid-render), accumulators
use canonical sparse initialization, DOM lookups stay null-guarded, and
length-checked constructions carry named invariants.
* feat(tooling): enforce noUncheckedIndexedAccess in the core and ui lanes
Flips the flag on for tsconfig.core.json and tsconfig.ui.json (covering
src, all packages including the two deferred ones, and ui) and retires
the strict-ratchet lane wholesale: config, script, projects reference,
check/CI wiring, changed-lane routing, and sync test. The assertion-ban
oxlint override stays as an independent surface.
