- the constants name one bound each
- resolveLocalInstant reuses an instance a caller already holds instead of
copying it per candidate date
- the README entry-point paragraph drops its duplicate sentence
- the note replaces an invalid reason for the zone mapper and a
review-vantage sentence with the facts
- isFourDigitYearInstant is the shared predicate; Schedule's five
representability guards delegate to it instead of re-deriving the range
- the reason for a field outside its calendar range is out_of_range in
both strict parsers, so year 0 no longer reads as a wall-clock problem
- LocalCalendarTime.year states which entries reject year 0 and which
accept it at the UTC edges
- the note counts the five removed expressions and records that the zone
mapper is a silent catch-all; the Chinese side mirrors the English
paragraph order
- the README no longer claims resolveLocalInstant answers a string
- the accepted-input table records the widened plain date-time input
- canonicalClientTimeZone is named as the wire-boundary form rather than
left outside a blanket claim about discriminated results
- the non-canonical scheduledAt diagnostic has one source
- resolveLocalInstant accepts an already-built plain date-time, so the
recurrence walkers drop their private copy of the disambiguation and
round-trip decision
- decodeInstant narrows its field explicitly and drops the assertion
- the model-input exhaustiveness fallbacks throw a plain Error, matching
the timing-change fallback, while durable decoding keeps
ScheduleLogError
- the library module and README state which entries answer a
discriminated result
- the guarded ignore covers only the uncovered branch point
- the note says two consumers copied the profile, counts copies rather
than consumers, and scopes the platform cause to the unsupported
rejection
- the note names the remaining instant-profile copy in session-format
time-context kept a third copy of the IANA profile: its own regular
expression and its own Intl canonicalization in request-zone.ts. It now
reads a request's reported zone through canonicalizeTimeZone and keeps
its stricter policy — an alias whose canonical answer differs from the
reported value is still refused, with the same three diagnostics.
The shared rejection now carries the platform error it caught, so the
unsupported refusal keeps its cause in both time-context and Schedule.
The shipped Web bundle gives every live root Agent the Host Schedule
tools, and every other scenario in this file that boots the shipped
base/web bundle with the minimal preset expects ['bash',
...SCHEDULE_TOOLS]. This case uses the same bootWeb composition and adds
only a settings patch, which does not change tool composition, so its
['bash'] expectation failed on every PR.
The shipped Web composition now puts four Schedule tools on every live root
Agent, which pushes a full-preset `cordis_inspect_query` answer past the
12,500-token inline budget, so the spill policy retains its head and tail with
a spill path instead of the complete JSON.
Refreshed owners publish a v4 generation, so every scenario that names
another owner's Session fixture must name that owner's highest generation.
The shipped Web composition also puts the Host Schedule tools on every live
root Agent, so the minimal-preset tool-table assertions include them.
The Web bundle now mounts `time-context` and `schedule` and leaves the
`ui-schedule` client row enabled, so the Automation tasks page, the Session
reminder catalog, and the `schedule_*` tools work without an overlay. Both
packages join the bundle's declared dependencies, and the example overlay is
deleted: re-inserting its rows would mount both Host plugins twice.
The fixed-rate floor drops from 300 to 60 seconds across the Host decoder,
the create and update validation, the model-visible tool schema, and the
client's per-unit floors.
Web recorded sessions and header sidecars are refreshed for the readings
`time-context` now appends; `normalizeWebSessionVolatiles` tokenizes their
sampled instant, browser zone, and elapsed duration so replay stays stable.
- Instant entries that take already-validated values throw on violated
preconditions; the module header said every entry answers a result.
- formatFourDigitYearInstant also refuses a non-integral epoch.
- Schedule's reason mappers switch over the closed unions with a never
default, so a new library reason fails the build instead of falling
through to a catch-all.
- The Agent Note claimed the schedule suites pin these diagnostics; they
assert codes only, and the note now records the added message assertion.
A calendar date the month does not have keeps the at diagnostic; only an
out-of-range clock field keeps the local at one. The extraction had
collapsed both onto the local at wording.
Move Schedule's strict RFC 3339 instant, local-calendar, and IANA zone
parsing into @deepseek-ai/dsh-util-time, and have Schedule consume it.
The shared entries answer discriminated results and own no failure
vocabulary; Schedule keeps its error codes, diagnostics, durable record
shapes, and all recurrence arithmetic.
The composer's paste path used RangeSelection.insertText, which keeps every
pasted newline inside one text node. In that document the browser reports the
collapsed caret's geometry as the whole block's, so deleting the draft's last
line moved the capped scrollport to the draft's head while the caret stayed at
its end.
RangeSelection.insertRawText splits the input into text and line-break nodes,
the document a typed draft already has.
A summary's fenced code block pins its banner at the scrollport top, where the
pinned compaction header already sits, so the header covers the banner's Copy
control and a click there collapses the summary. Stop the banner below the
header's band and square the pinned header's corners so its opaque fill masks
the scrolling prose at the edges.
Both blocks render their expanded body uncapped, so a long reasoning chain or
compaction summary carried the disclosure header off the viewport and buried
the toggle. Pin each header to the conversation scroll container while the
block is open, keeping the rule scoped to these two blocks so the capped tool
rows do not stack headers at the top of the viewport.
The compaction header also overrides its stacking rank above the summary's
code-block banner and keeps an opaque hover fill.
The late-listener case now carries a pending generation change, so its
assertions depend on the disposed guard, and the silent-generation cases assert
their tick counts.
A notification repeating the published generation no longer withdraws displayed
options, so the invalidation observable and the README describe the same rule.
The composer seat hides on a cleared catalog while the slash picker keeps its
failure and retry, and the disposed guard that both callers already provide is
gone from the tick.
The package README, its Chinese pair, and the catalog spec now state both
public observables and pin the generation-change tick that withdraws displayed
options.
Publishing the result of a read the picker waits for no longer closes that
picker, so a failed read keeps its error and retry state; a catalog
notification or generation change still withdraws stale options. Correct the
action-history rationale in the Auto review note and cover both paths with the
real popup controller.
Availability now follows the Session projection alone, so a failed catalog
read surfaces through the picker's own retry instead of hiding the command.
Type-guard the persisted denial reason before trimming it, and record in the
Auto review note why the reviewer reads the complete action history.
Loader settlement no longer rejects on a failed optional entry, so the Auto
rollback scenario reads the conflict from the startup audit. Refresh the Auto
denial golden for the merged composer chrome.
Register the permission catalog default response so the web roster boot
matches every Remote call it makes, and carry the deprecation rule for the
session history read this reviewer needs with the deferred migration
recorded in the package README.
Resolve 36 conflicts across source, bilingual READMEs, and generated
documentation: keep master's command-identity and popup APIs, RemoteMock
test infrastructure, and regenerated catalogs; keep the Auto review
catalog, slot, dismissal, and snapshot content.
The managed-scope disposal cases in subprocess-local and bash-sandbox run past
the 5000ms default when this lane's four partitions, six workers per gate, and
sibling gates share one host. The same commit stays inside the budget on the
in-house pool: the hosted image exceeded it (run 34449848541) and the lane is
green with 90000 (run 34457655892). The Windows coverage lane already grants
the same budget for the same class of teardown.
(cherry picked from commit bf3468adf0947a51f09d4a1aaae2865f3e50a2f2)
The ACP plugin-config case drove a real dispose with 150ms graces. On the hosted
image the SIGKILL escalation landed while the managed scope could not take the
signal — systemctl failed the kill ("Failed to send signal SIGKILL to auxiliary
processes: Invalid argument") — so the run reported a teardown failure the
configuration never asked for. The case asserts that config graces reach the
real run, so the values only need to clear the host's scope handling. Its mock
refuses stdin EOF and SIGTERM by design, so the case waits out both graces and
carries its own budget for that fixed cost.
Both illegal-UTF-8 residual cases relied on a plain sched_yield leaving each
write as its own read chunk. A starved reader coalesces hundreds of bytes into
one chunk, which inflates the measured peak for correct accounting too: the
hosted image measured 2563 against the 2048 bound. Pace with a real sleep so
each byte is its own chunk everywhere, and keep each payload above the bound
(3200 bytes; 1100 CESU-8 sequences) so a raw-byte undercount still flushes past
it instead of only at EOF.
(cherry picked from commit 30e0c3ff0415befe3652a55611df8a18d82c18bb)
`disposal contains a spawn-failure rejection that races teardown` asserted one
winner of the race. A bootstrap that publishes its pre-exec failure rejects with
that failure, and a teardown that stops the bootstrap first settles as the
requested termination — the recorded failure only outranks the stop when it was
published before the stop landed. Assert the contract instead: a rejection
carries the failure, a stop carries SIGTERM, which only the Linux scope records
because the win32 job owner and the fallback launcher both reject the start.
The two terminal cases that drive a mocked PTY exit state why they pin
`internals.platform`: under the host's native scope the mocked exit races the
scope bootstrap and fails as `terminal scope exited before its bootstrap
consumed the launch request`.
(cherry picked from commit fee113eb9a5632555304858d2bf6355588a3aaf0)
The failover leg runs this suite on pools this repository does not own, and the
coverage lanes failed there on host properties their cases never named: the
containment a mocked PTY exit raced, the grace a managed scope needs before it
can take a SIGKILL, reader coalescing in the illegal-UTF-8 residual cases, and
a Windows image that refuses the folder dialog class. Record what each case
pins now, what was rejected, and the runs that measured it, so the next pool
change starts from those host dependencies instead of re-diagnosing them.
The managed-scope disposal cases in subprocess-local and bash-sandbox run past
the 5000ms default when this lane's four partitions, six workers per gate, and
sibling gates share one host. The same commit stays inside the budget on the
in-house pool: the hosted image exceeded it (run 34449848541) and the lane is
green with 90000 (run 34457655892). The Windows coverage lane already grants
the same budget for the same class of teardown.
The ACP plugin-config case drove a real dispose with 150ms graces. On the hosted
image the SIGKILL escalation landed while the managed scope could not take the
signal — systemctl failed the kill ("Failed to send signal SIGKILL to auxiliary
processes: Invalid argument") — so the run reported a teardown failure the
configuration never asked for. The case asserts that config graces reach the
real run, so the values only need to clear the host's scope handling. Its mock
refuses stdin EOF and SIGTERM by design, so the case waits out both graces and
carries its own budget for that fixed cost.
Both illegal-UTF-8 residual cases relied on a plain sched_yield leaving each
write as its own read chunk. A starved reader coalesces hundreds of bytes into
one chunk, which inflates the measured peak for correct accounting too: the
hosted image measured 2563 against the 2048 bound. Pace with a real sleep so
each byte is its own chunk everywhere, and keep each payload above the bound
(3200 bytes; 1100 CESU-8 sequences) so a raw-byte undercount still flushes past
it instead of only at EOF.
`disposal contains a spawn-failure rejection that races teardown` asserted one
winner of the race. A bootstrap that publishes its pre-exec failure rejects with
that failure, and a teardown that stops the bootstrap first settles as the
requested termination — the recorded failure only outranks the stop when it was
published before the stop landed. Assert the contract instead: a rejection
carries the failure, a stop carries SIGTERM, which only the Linux scope records
because the win32 job owner and the fallback launcher both reject the start.
The two terminal cases that drive a mocked PTY exit state why they pin
`internals.platform`: under the host's native scope the mocked exit races the
scope bootstrap and fails as `terminal scope exited before its bootstrap
consumed the launch request`.
A Windows Server image can answer CoCreateInstance(CLSID_FileOpenDialog)
with CLASS_E_CLASSNOTAVAILABLE, so the win32 COM smoke has no dialog to
open and the POSIX rejection expectation is the only proof that image can
produce. Gate the pair on a PowerShell activation probe taken outside the
code under test, so a GUID or vtable regression in src still fails the
smoke instead of skipping it.
The failover runbook, the node-compatibility note, the native-Windows note,
and the serial-reference note record the third variable value's scope and
exclusions (bilingual, pairing re-recorded); a new implemented process note
owns the value's participating set, the dormant-by-default canary posture,
and the walker-fix rationale. .github/AGENTS.md points at both notes.
Node's internal fs.glob, from some 24.x releases, lstat-probes
<matched>/<next segment> while expanding a ** pattern over a symlinked
file and throws ENOTDIR instead of skipping (observed on node 24.13.0
scanning snapshots/acp/image-compaction's symlinked
system-prompt.expected.md, which failed verify-md-wrap inside
check:ci:static). uniqueRepoFiles now walks with dirent types, never
probing a path under a file, and mirrors fs.glob's dot:false and
follow:false semantics: the four repo-files gate pattern sets enumerate
identically. Broken or cyclic symlinks fail loudly in the caller's
realpathSync, and pattern forms the walker does not model (metacharacters,
a trailing **, empty segments) are rejected up front. repo-files.spec.ts
pins the symlink, broken-link, and rejection behavior.
The per-platform failover repository variables (DSH_CI_FAILOVER_LINUX and
DSH_CI_FAILOVER_WINDOWS) accept a third value alongside the existing
selfhosted routing. Set by a repository writer, the value retargets the
always-on pull-request and master lanes onto the matching hosted-runner
labels (16-core pools map to the 16-vCPU tier, standard legs to 4-vCPU);
unset or any other value keeps today's runners, so the leg is dormant until
a writer opts in - a canary path for capacity fallback, never the default.
Participating jobs: the three enterprise Linux workers, the node-compat
legs, the four Windows lanes, and the all-checks-passed verdict in ci.yml;
expected-filenames; the sandbox bwrap leg; and the two manual benchmark
matrices (4-32 vCPU tiers map to equivalents; 64/96-core rows keep the
fleet labels). Everything pinned to its own runner by spec, credential, or
host capability keeps it under every value, and the release-shaped chains
stay entirely hosted. request-review.yml was retired on master while this
work was in flight and the branch takes the deletion.
Specs pin the new branches: ci-workflow.spec.ts asserts the labels appear
in the extended selectors, and ci-compatible-selfhosted.spec.ts evaluates
the routing mode.
Second review round (ds-review-bot v4p/v5/v6): the exit cleanup claim that
collectors unlink their spill files on dispose was wrong — completed spill
files are retained (seal() only closes, discardSpill() is the only unlink),
so the exit removal only ever applies to directories that never spilled.
Reword the JSDoc, README, and Agent Note accordingly, drop the redundant
readdirSync precheck (rmdirSync ENOTEMPTY is caught anyway), and recount the
v8-ignore window. The spawn.spec default-spill test now records the default
per-process directory it created and removes it in afterAll, so a completed
spill file is never left behind by the unit suite.
The Agent Note now links the 2026-07-17 local-spill-startup-cleanup note it
builds on, and the privateSpillDir JSDoc documents the empty-dir exit
removal alongside the directory's creation contract.
Review (ds-review-bot) found the exit-time deletion of per-process spill
roots conflicts with the documented retention decision: spill artifacts are
model-visible locators that resumed or forked sessions may still reference
(2026-07-17-local-spill-startup-cleanup), and an exception thrown from an
'exit' listener can change the process exit code.
- dsh-spill-local: revert the exit handler entirely; its default root is
already owned by the package's 30-day startup sweep.
- dsh-subprocess-local/spawn: remove the per-process spill dir at a
JavaScript-observable exit only while it is EMPTY (collectors unlink
their spill files on dispose), best-effort with a named swallow so a
Windows-held handle never changes the exit code; dirs holding spill
files keep them for external cleanup.
- Document the empty-dir exit removal in the package README and the Agent
Note, and re-record both bilingual pairs.
The exit listener runs after the coverage dump, so its body can never be
measured by the unit coverage lane; mark it v8-ignore with the reason, as
with other process-exit-only code paths.
Full-template residue histogram on the CI host surfaced four more normal-exit
leaks below the earlier cutoff: tool-bash/tool-pwsh tools.spec (module spill
dir / per-test homes), bash-sandbox sandbox.spec (module spill dir and the
read-only denial root), and fs-sandbox fs-sandbox.spec (inline tmp dir). All
four now remove what they create at the same teardown points as their
neighbors.
Spec files that create /tmp/dsh-* directories via mkdtemp now track and
delete them in afterEach/afterAll; module-scope fixture dirs (executor
spill dirs) are removed in afterAll. The file list came from the
observed-residue inventory on the self-hosted CI host: only specs whose
dirs actually accumulated were leak sources (issue #3134), superseding
the kept-but-unmerged CI sweep branch per the #3233 review decision.
Product per-process spill roots (dsh-subprocess-local spawn,
dsh-spill-local store) register a process-exit handler that removes the
memoized dir, so processes that used the spawn/spill path clean up on
normal exit. A SIGKILLed process cannot run in-process teardown; the
machine-side timer remains the backstop for that path.
Agent Note: .agents/notes/implemented/process/2026-08-28-test-temp-dir-self-cleanup.md
Review feedback: the dsh-terminal-bash product default is 30s, not 300s;
the raised value bounds one send plus the complete startup sequence, so it
must cover the same cold start the tool deadline does. The vitest case
budget stays at its pre-existing 120s: the case-level timeout overrides
the lane --testTimeout, so syncing it to the plugin deadline would make a
stalled partition wait 300s instead of 120s.
The first call pays the full pwsh cold-start latency (spawn + .NET +
PSReadLine + Defender) inside the tool deadline. A 60s bound on the fully
loaded self-hosted Windows pool is exceeded often enough to reset the
session mid-test: two master CI runs (2026-09-01, runs 33524764567 and
33534262413) each failed at ~62s with the second call observing a fresh
session (cwd back at root, env empty). 300s matches the product default
so cold start no longer races the budget.
The hostFrameParseCeiling JSDoc and one load-gate test comment still quoted
the pre-16x derivation (~29 MiB for a ~300 MiB heap, ~14 MiB for a 128 MiB
old space). With HOST_PARSE_WORST_CASE_MULTIPLE = 16 the same hosts derive
~14 MiB and ~7 MiB (floor((176-64)/16)); protocol.spec.ts pins the 304 MiB
case at 15 MiB. Comment-only correction, no behavior change.
The #3289 merge-forward dropped four typescript/no-unnecessary-condition
suppressions from index.ts (boot-write-failure fake child stdin, the
admit()-closure logsTruncated recheck, and both settled rechecks whose
guards flip mid-wait), turning the lint:contracts-ready gate red. Re-add
them with their reasons. The merge also carried a ptc-python-turn session
fixture that was not in canonical packed layout; migrate-packed-session-fixtures
re-writes it so session-fixture-layout passes.
render() loaded the workspace manifests once per external dependency
name through workspaceLinkedManifest, an O(names x manifests) file
read on the cold path; on the loaded self-hosted Windows host with
coverage instrumentation the freshness spec crossed Vitest's default
5000ms budget and failed the serial-windows standby gate four times in
a week. Load the manifests once in render() and thread the map through
the collectors instead.
The serial-windows lane also ran the coverage inventory at the strictest
budget of any lane: add DSH_COVERAGE_TEST_TIMEOUT_MS=90000 to match the
pull-request windows-coverage lane, pinned by ci-workflow.spec.ts.
Review findings on the CPython backend: a child that never reads fd 3 leaves
the reply pipe full forever, so the drain loop waits on 'drain' while every
call frame it keeps sending resolves a binding and queues another reply —
the backlog (and the binding results it pins) would grow until the wall
clock. sendReply now caps the pending backlog at MAX_PENDING_REPLIES and
settles the run as worker-exit past it, mirroring the frame cap; a child
flooding calls against a binding that never settles would otherwise bypass
that cap (pendingReplies grows only after the await), so the dispatcher
counts in-flight binding calls before dispatch and releases the slot in the
async body's finally, capping outstanding closures at the same bound. The
drain also compacts its consumed prefix (replyQueue.splice(0, head)) once
head reaches the bound, so a drain that stays alive without emptying cannot
grow the backing store linearly with cumulative throughput.
The completion-value meter counted lone surrogates with
_SURROGATE.findall(folded), materializing one single-character string per
surrogate: a surrogate-dense value near the budget (millions of surrogates,
each serializing to six bytes) allocated millions of objects before the meter
returned, defeating the meter's counting-without-building contract. The count
is now the length difference between folded and the without string the meter
already computes; a standalone equivalence check confirms it matches findall
across lone-high, lone-low, paired, astral, and mixed cases.
validateBindings read namespace.global/errorClass.name/memberNameProperty
several times and retained the original errorClass object for the boot
frame, whose JSON.stringify re-read it after validation: a stateful getter
could throw or change between the two stages, turning the seam-misuse
rejection into a worker-exit or injecting an unvalidated name. Each field is
now read once into a plain value and the bindings map stores a plain
{ name, memberNameProperty } copy, so validation and the boot frame see
identical values.
Regression tests: a hostile child floods 5000 sequential valid calls without
reading fd 3 and the run settles worker-exit with the reply-queue message
before maxWallMs; a 3,000,000-surrogate completion succeeds at an
18,000,002-byte budget and reports output-limit one byte under; a 5000-call
flood against a never-settling binding settles worker-exit with the
call-backlog message; getter-backed namespace metadata that throws or
changes on a second read boots and runs with each field read exactly once; a
two-wave flood whose replies exceed the writable high-water mark drives the
drain past the compaction bound mid-delivery and verifies all 1524 replies
arrive. README Known Limitations gains the reply-backlog and call-backlog
bounds (en/zh, pairing re-recorded); a new Agent Note registers the findings.
The capability-seams graph derives its implementation lists from
SERVICE_ROLES in scripts/gen-doc-graphs.ts, which still listed only the
worker-thread backend. Add experimental-code-runtime-python so the
generated graph and table match the registered ctx.codeRuntime
implementations; regenerate docs/capability-seams.md, sync the zh pair,
and re-record the i18n pairing.
Review findings on the CPython backend: an explicit pythonBin path bypassed
the load-time checks (missing/non-executable/directory paths surfaced only
as a run-time worker-exit); a throwing binding member accessor escaped the
fd-3 data callback and terminated the host; the reply drain waited on
'drain' alone, so a pipe destroyed under the wait hung forever; and two
staging-leak assertions diffed a global tmpdir that parallel workers can
perturb.
resolvePythonBin now applies the same accessSync(X_OK) + isFile check to
explicit paths (resolved against the host CWD), and the load error message
distinguishes 'is not an executable regular file' from 'does not resolve on
PATH'. validateBindings snapshots callables into a plain record during run()'s
synchronous validation, turning an accessor throw into the seam-misuse
rejection and fixing the key set the boot frame and dispatch share. The reply
drain waits on drain/close/error together and short-circuits on
proto.destroyed. The staging-leak assertions check the exact paths this test
file staged (recorded by the mocked mkdtempSync) instead of a tmpdir diff.
docs(code-runtime-python): add the alternatives section to the hardening note
docs(config-catalog): refresh the code-runtime-python Config source line
test(code-runtime-python): cover the async spawn-error worker-exit path
Chinesezjc