Repository navigation
Conversation
Size Report
Startup median (7 runs, lower is better):
|
There was a problem hiding this comment.
All reported issues were addressed across 12 files
You’re at about 99% of the monthly reviewed-line limit. You may want to disable incremental reviews to conserve quota. Reviews will continue until that limit is exceeded. If you need help avoiding interruptions, please contact contact@cubic.dev.
Reply with feedback, questions, or to request a fix.
View guided diff | Turn on auto-fix | Re-trigger cubic
Apex Sentinel · ✅ ApproveSince the last review: 0 new · 0 still open · 1 no longer raised No open findings. 1 finding below 💬 Talking to SentinelIn a finding's thread, reply with:
On this conversation, comment Earlier reviewsReview of
|
| Where | Finding |
|---|---|
packages/host-kit/src/internal/exec.ts:309 |
🆕 An output-limit kill that fails to stop the child now hangs runCmd forever — the deadline is skipped |
2 findings below minor hidden (user_settings.min_severity).
Reviewed with this repository's own Sentinel configuration (1cb2f242aea5).
Review of 0399650f · 17.11.2026 12:56 UTC
Apex Sentinel · ✅ Approve
Since the last review: 0 new · 0 still open · 0 no longer raised
No open findings.
1 finding below minor hidden (user_settings.min_severity).
Reviewed with this repository's own Sentinel configuration (fdf595127a75).
Review of 472f2c2d · 17.11.2026 11:44 UTC
Apex Sentinel · ✅ Approve
Since the last review: 0 new · 0 still open · 0 no longer raised
No open findings.
Reviewed with this repository's own Sentinel configuration (fce2fa6247c5).
Review of 00e8e055 · 17.11.2026 09:38 UTC
Apex Sentinel · ✅ Approve
Since the last review: 0 new · 0 still open · 1 no longer raised
No open findings.
2 findings below minor hidden (user_settings.min_severity).
Reviewed with this repository's own Sentinel configuration (e0ab0c255c3d).
Review of 34b1daa0 · 17.11.2026 08:31 UTC
Apex Sentinel · ✅ Approve
Since the last review: 0 new · 1 still open · 0 no longer raised
1 finding(s): 1 minor.
🟡 Minor (1)
| Where | Finding |
|---|---|
packages/host-kit/src/internal/exec.ts:264 |
Output-limit breach throws even for callers that pass allowFailure: true |
Reviewed with this repository's own Sentinel configuration (e0ab0c255c3d).
Review of 372a542f · 17.11.2026 07:17 UTC
Apex Sentinel · ✅ Approve
1 finding(s): 1 minor.
🟡 Minor (1)
| Where | Finding |
|---|---|
packages/host-kit/src/internal/exec.ts:264 |
Output-limit breach throws even for callers that pass allowFailure: true |
1 finding below minor hidden (user_settings.min_severity).
Reviewed with this repository's own Sentinel configuration (e0ab0c255c3d).
There was a problem hiding this comment.
Apex Sentinel: 1 new finding on 372a542. The verdict and the full list are in the summary comment.
|
This PR is ready at 34b1daa. The code looks correct, and all 14 checks pass at that commit. There are no conflicts. The default path (maxBuffer omitted) is covered by the real-subprocess tests in exec.test.ts, so no device run is needed. Nothing blocks merge except order: the PR targets the stacked base fix/a1-extraction-unused-import-20261116, so it merges after that layer. Not blocking, and you can take or leave these: (1) the precedence abort > timeout > output_limit lives in both commandSettlementReason and the guard in settle, so branching settle on the final reason would give the ladder one owner; (2) commandSettlementReason and normalizeTimeoutMs are not tracing code and only exec.ts uses them, so they fit better in exec.ts or an exec-settlement module, unless a size gate forced the move; (3) bindCommandLifecycle now sits between the CommandKillSettlement doc block and its type, and execResult lost its doc line; (4) validation-case.ts re-exports encodeValidationCase, so validation-arbitraries-cli.ts has two spellings of one symbol, and it would be cleaner to import from @agent-device/test-kit and drop the re-export; (5) CASE_GENERATION_INPUTS lists the test-kit barrel, which changes for unrelated helpers, so listing only the two generator files is enough; (6) provider-webdriver.ts is the only production caller that passes maxBuffer (10 MiB), so cloud WebDriver host commands above that now fail with command_output_limit, which is worth naming in the PR body, and runCmdSync could map ENOBUFS to the same details. On the other review threads: the threads on background maxBuffer, retry block, runScript waiver and coversCommand matching are fixed at this head, so they can be resolved. The allowFailure thread does not apply: allowFailure covers non-zero exits only, timeouts and sync buffer overflow already throw regardless, and an output-limit breach is a resource violation, so throwing is consistent. |
|
Reviewed the six optional notes against
The output-limit error now retains bounded captured prefixes. The five tracked review threads are resolved. New-head hosted checks and parent-layer acceptance remain separate from the prior |
|
I reviewed 00e8e05 against the earlier review at 34b1daa (#4621 (comment)). I found no new problems in the delta. The output limit error now carries the bounded stdout and stderr prefixes, and the trace reports |
|
Re the output-size question in #4621 (comment): CLI and daemon wire errors pass through |
472f2c2 to
0399650
Compare
0399650 to
e60df0c
Compare
There was a problem hiding this comment.
Apex Sentinel: 1 new finding on e60df0c. The verdict and the full list are in the summary comment.
e60df0c to
ccbf9ed
Compare
|
The rebase requested in #4621 (comment) is published: The sole new commit is the deadline termination fix: three production lines and a real-child regression. It retries a failed first kill at the existing deadline while retaining the earlier output-limit error. The controlled old-source run failed both The remaining gate is the Chief of Staff review of this new commit. I have not changed readiness labels. All four stack heads will stay fixed while that review completes; no further code change or rebase is currently needed. |
|
I found no problems in ccbf9ed. The conflict from the earlier review (#4621 (comment)) is resolved. The output-limit kill no longer skips the deadline: exec.ts:309-314 re-requests the kill and keeps didTimeout false, so a failed first kill cannot hang runCmd. That fixes the open thread at #4621 (comment), so you can resolve it. CI is green: 15 checks, none failing (13 success, 1 skipped, 1 neutral). The test stubs child.kill to return false without sending the signal. That reaches the retry branch, because signalProcessTree ignores the return value. #4579 needs to land first, and nothing else stops this PR. |
* refactor(test): colocate Maestro fuzz coverage with parser * fix(host-kit): bound captured command output * fix(host-kit): count raw output bytes before decoding * refactor(host-kit): isolate command settlement classification * fix(host-kit): preserve timeout and eager closure * fix(test-kit): cover complete Maestro command tokens * style(test): format Maestro coverage assertion * fix(host-kit): retain bounded output on capture limit errors * test: coordinate bounded output regression * fix(host-kit): retry output-limit termination at deadline
…ent vocabulary 0.22 (#4621 bounded output) and main (#4777 startup progress) each grew packages/host-kit/src/internal/exec.ts, so the merged file passed its legacy 1032-line cap and lint failed. Re-home the kill/settle ordering — signalProcessGroupBestEffort, bindCommandLifecycle, the kill settlement, the abort watch, killProcessTree and destroyCommandStreams — into exec-trace.ts, which already owns the exec settlement vocabulary and sits in every closure that reaches exec.ts, so no eager closure grows. exec.ts now fits the 1000-line rule and its legacy cap entry is removed.
* refactor(test): colocate Maestro fuzz coverage with parser * fix(host-kit): bound captured command output * fix(host-kit): count raw output bytes before decoding * refactor(host-kit): isolate command settlement classification * fix(host-kit): preserve timeout and eager closure * fix(test-kit): cover complete Maestro command tokens * style(test): format Maestro coverage assertion * fix(host-kit): retain bounded output on capture limit errors * test: coordinate bounded output regression * fix(host-kit): retry output-limit termination at deadline
…ent vocabulary 0.22 (#4621 bounded output) and main (#4777 startup progress) each grew packages/host-kit/src/internal/exec.ts, so the merged file passed its legacy 1032-line cap and lint failed. Re-home the kill/settle ordering — signalProcessGroupBestEffort, bindCommandLifecycle, the kill settlement, the abort watch, killProcessTree and destroyCommandStreams — into exec-trace.ts, which already owns the exec settlement vocabulary and sits in every closure that reaches exec.ts, so no eager closure grows. exec.ts now fits the 1000-line rule and its legacy cap entry is removed.
* refactor(test): colocate Maestro fuzz coverage with parser * fix(host-kit): bound captured command output * fix(host-kit): count raw output bytes before decoding * refactor(host-kit): isolate command settlement classification * fix(host-kit): preserve timeout and eager closure * fix(test-kit): cover complete Maestro command tokens * style(test): format Maestro coverage assertion * fix(host-kit): retain bounded output on capture limit errors * test: coordinate bounded output regression * fix(host-kit): retry output-limit termination at deadline
…ent vocabulary 0.22 (#4621 bounded output) and main (#4777 startup progress) each grew packages/host-kit/src/internal/exec.ts, so the merged file passed its legacy 1032-line cap and lint failed. Re-home the kill/settle ordering — signalProcessGroupBestEffort, bindCommandLifecycle, the kill settlement, the abort watch, killProcessTree and destroyCommandStreams — into exec-trace.ts, which already owns the exec settlement vocabulary and sits in every closure that reaches exec.ts, so no eager closure grows. exec.ts now fits the 1000-line rule and its legacy cap entry is removed.
* refactor(test): colocate Maestro fuzz coverage with parser * fix(host-kit): bound captured command output * fix(host-kit): count raw output bytes before decoding * refactor(host-kit): isolate command settlement classification * fix(host-kit): preserve timeout and eager closure * fix(test-kit): cover complete Maestro command tokens * style(test): format Maestro coverage assertion * fix(host-kit): retain bounded output on capture limit errors * test: coordinate bounded output regression * fix(host-kit): retry output-limit termination at deadline
…ent vocabulary 0.22 (#4621 bounded output) and main (#4777 startup progress) each grew packages/host-kit/src/internal/exec.ts, so the merged file passed its legacy 1032-line cap and lint failed. Re-home the kill/settle ordering — signalProcessGroupBestEffort, bindCommandLifecycle, the kill settlement, the abort watch, killProcessTree and destroyCommandStreams — into exec-trace.ts, which already owns the exec settlement vocabulary and sits in every closure that reaches exec.ts, so no eager closure grows. exec.ts now fits the 1000-line rule and its legacy cap entry is removed.
* refactor(test): colocate Maestro fuzz coverage with parser * fix(host-kit): bound captured command output * fix(host-kit): count raw output bytes before decoding * refactor(host-kit): isolate command settlement classification * fix(host-kit): preserve timeout and eager closure * fix(test-kit): cover complete Maestro command tokens * style(test): format Maestro coverage assertion * fix(host-kit): retain bounded output on capture limit errors * test: coordinate bounded output regression * fix(host-kit): retry output-limit termination at deadline
…ent vocabulary 0.22 (#4621 bounded output) and main (#4777 startup progress) each grew packages/host-kit/src/internal/exec.ts, so the merged file passed its legacy 1032-line cap and lint failed. Re-home the kill/settle ordering — signalProcessGroupBestEffort, bindCommandLifecycle, the kill settlement, the abort watch, killProcessTree and destroyCommandStreams — into exec-trace.ts, which already owns the exec settlement vocabulary and sits in every closure that reaches exec.ts, so no eager closure grows. exec.ts now fits the 1000-line rule and its legacy cap entry is removed.
Summary
Bound captured host-command output by raw bytes per stream while preserving decoding, timeout and cancellation behavior. Cloud WebDriver host commands use a 10 MiB per-stream cap. Typed
command_output_limiterrors retain bounded stdout/stderr, including withallowFailure.The deadline now retries termination when the first output-limit kill fails, preserving the earlier output-limit reason. Move fuzz support to test-kit and keep fixed-seed Maestro completeness beside the parser. This 12-file, 477-gross-line prerequisite supports #4617 and targets #4579.
Validation
Published head:
ccbf9edfc5e1bc34e21e459bb875617b1d3ee6c3, based on A1c6103613. Pinned frozen install/build and the exact affected gate passed: 42 checks passed, 25 skipped, unchanged checkout. Compared with the last reviewede60df0c, nine patches are unchanged;ccbf9edis the sole new commit. Its change is three production lines plus a regression. The new-commit Chief of Staff review remains pending.The real-process deadline regression failed on the old source for both
allowFailuresettings and passed with the fix; all 32 executor cases passed. Root reviewed the fix and tests. Existing prefix regressions cover both streams.The prior public conflict is repaired by the pushed rebase. Hosted results at this head completed: 13 successful, one skipped and one neutral. Sentinel reports no open findings and no longer raises the deadline concern; one below-minor finding is hidden by its configured threshold. All six review threads are resolved. Land #4579 first, after Chief of Staff clearance for this commit.
Open in [code]smith · Need help on this PR? Tag
@codesmith-botwith what you need.