Node 26.0.0 removed --experimental-transform-types, so the native
source-launch chain cannot start anywhere on that line, and strip-only
mode rejects the vendored syntax (parameter properties, decorators,
runtime enums/namespaces). Switch bin/dsh, the root dsh/demo:tui/
demo:web scripts, and the Code Mode TUI overlay to node --import
tsx/esm: one launch vector across the whole engines range, ~0.4s faster
than the full tsx default (the CJS hook stays off; the graph is
ESM-only).
Delete scripts/tspath-loader.ts and apps/cli/src/tsconfig-paths-loader.ts:
tsx owns both transformation and tsconfig paths projection. Add
dsh-source-launch-smoke to the node-compat gates so the 22.19/26 matrix
executes the real launch vector; no CI job did, which is how the Node 26
breakage shipped silently.
Supersedes the native-TypeScript-source-launch Agent Note (new note
records the profiling evidence and rejected alternatives).
Both packages import @deepseek-ai/dsh-llm at runtime (createUserMessage)
but declared it only in devDependencies; any resolver that honors
declared runtime dependencies resolves the import to a stale or missing
artifact. Found by the (since removed) source-launch declared-dependency
check.
The self-referential inspect tool surfaces Session's public API in its
transcript; the construction-boundary field added in bc8512d17 now
appears there. Keyless refresh; only the affected scenario re-recorded.
Resolutions: regenerate the conflicted generated docs (cordis services
catalog, event-producer-consumer, module-graph); take master's
packages/README pair and re-insert the telemetry row on both sides;
re-record the README and session-doc translation pairs.
Review finding, pinned red-first: the SDK accepts
processor.maxExportBatchSize <= 0 (or fractional), but its shutdown
drain then splices empty batches without consuming the queue —
disposing telemetry hangs forever whenever records are queued. The
constructor now rejects a non-positive-integer batch size before
building the SDK processor, per the misconfiguration-fails-loud rule;
everything else in the processor block remains the SDK's verbatim
passthrough.
Review finding: synthetic turn closers written by SessionPersistence
.load() at crash-resume sit below firstLiveSeq and are never exported,
leaving the remote turn unbalanced. Kept deliberately — exporting a
synthetic closer cannot complete a turn whose real tail records died
in the crashed process's queue; it can only make an incomplete turn
look closed. The OTel README now states the receiver rule (a
never-closed turn on a resumed stream marks the previous process dying
inside it; a later clean shutdown marker attests only to the resumed
process's exit), and the revival Agent Note records why the repair
suffix stays local.
The baseline columns pushed the two INSERT binding lists past the clone
threshold; one _headerBindings helper now owns the shared header prefix
(the duplication gate's finding).
A cursor-less adoption (process restart + resume, fork, seam-module
reload) replayed the session's full log from seq 0, re-exporting
history that already left the process — a resume re-billed its entire
stored log on every restart, and a fork re-shipped the parent's prefix
under the child's id, doubling query-time counts on OTLP backends with
no native ingest dedupe.
dsh-session now exposes the fact the constructor already validated but
discarded: Session.firstLiveSeq, the constructor-seed length — the
first seq appended in this process. header.seedLength cannot serve
here: it is the durable fork-lineage boundary, and a resumed session's
constructor seed is its full stored log while the header keeps the
original fork value (llm-replay and session-query-sqlite depend on
that meaning). Constructor seeds also never publish on the
session/event firehose, so adoption replaying them was inconsistent
with the system's own publication semantics.
Adoption's cursor-less fallback starts at firstLiveSeq; seed events
still feed the chunk projection, so mid-step continuations re-drop
after a resume. Fork streams are no longer self-contained: records now
carry session.seed_length (with the existing session.parent_id) so
receivers stitch the child's stream onto the parent's. Accepted cost,
consistent with at-most-once delivery and recorded in the revival
Agent Note: a resume no longer backfills records a previous process
failed to deliver — a deployment with that requirement needs the
deferred outbox, not replay.
Pinned red-first: seeded adoption exports nothing (assertion reversed
from the prior seed-readback test, obsolete behavior changed with its
test), resume-shaped seed rebuilds the projection without exporting,
and fork records carry the stitch attributes.
- knip: the subagent-inheritance resume fixture is a Loader-loaded entry
(the semantic-checkpoint-agent precedent), invisible to import analysis.
- coordinator coverage: a quiet-signal observation read pins the success
arm of observeQueuedAbort, and an approval-only baseline pins the other
arm of the seed-boundary guard's baseline disjunction.
Review fix (ds-review-bot on #623): the per-session-modes section still
described first-turn stamping — superseded by the header baseline — and its
formula lacked the middle rung. The formula now reads own post-seed knob
events ?? inherited SessionHeader baseline ?? composition default, and the
paragraph describes the synchronous capture into the creation-time header
with no first-turn timing window. Chinese counterpart updated in the same
commit.
Review fix (ds-review-bot on #623): the READMEs and JSDoc still stated the
pre-baseline formulas — resolve() outranking 'the session's last
sandbox/mode event', 'effective = explicit grant ?? fold(events) ??
deployment default', and the approval README's 'last approval/policy event'
opener — which contradict the shipped semantics for a delegated fork whose
seed tail differs from its header baseline. Every statement now names the
override chain (own post-seed switches ?? inherited header baseline): both
READMEs (both languages), resolve()'s JSDoc, the session-mode module and
event-declaration docs, the raw folds re-scoped as building blocks, and
the regenerated catalogs.
Review fix (ds-review-bot on #623): the ACP scenario runs at deployment
strength (the automation protocol has no session-scoped switch), so the
assembled-app path could not detect the delegation bypass itself. The new
keyless subagent-inheritance headless scenario closes that on the
semantic-checkpoint precedent: a seeded parent log carrying a real
sandbox/mode: read-only switch under a workspace-write deployment default
is resumed through the Loader-booted cli-demo app via a resume fixture
plugin and delegates through the real subagent tool; the child's real
write is denied by the real dsh-fs-sandbox fence (physical ENOENT
assertion), its persisted header carries the inherited baseline, and both
logs pin as expected outputs. Verified red: disabling the driver's capture
makes the scenario fail on the disk assertion (the child writes under the
deployment default).
Review fix (ds-review-bot on #623): matching baselines with differing
seedLength still resolve different policies — overrideOf folds own switches
past the boundary, so a stored seedLength: 1 marks event 0 as subsumed seed
history while a live seedLength: 0 lets the same event tighten the session;
adoption retained the stored header and a restart silently restored the
wide baseline. When either side carries a baseline, the seed boundary is
part of the policy identity and a mismatch rejects as an id collision.
Red-first in the shared coordinator contract (both backends).
Review fixes (ds-review-bot on #623):
- Persistence adoption compares the immutable policy baselines: onCreated's
ownerless claim and adoptLivePrefix retain the STORED header, so a
same-id live session with a conflicting baseline now rejects as a
collision instead of appending under read-only and resuming under the
stored danger-full-access.
- resolve() resolves the session override BEFORE applying an explicit
approved mode: the one-shot grant no longer bypasses the unconditional
durable-header validation.
- The approval narrator attributes positionally over the session's OWN
events (past the seed boundary): a fork child whose baseline delta has
no own override narrates 'inherited from the delegating session' instead
of misattributing a stale seed-carried switch to the user or the
operator.
Red-first: baseline-conflict adoption in the shared coordinator contract
(both backends), resolve-with-explicit-mode validation, and the fork-child
narration attribution case.
Review fixes (ds-review-bot on #623):
- overrideOf (both knobs) rejects a seedLength past the log end before
slicing: a malformed durable boundary would otherwise empty the
own-switch slice until the log outgrew it, letting a wide baseline
shadow a REAL later tightening. Malformed durable metadata fails loud,
never open.
- The session-query derived index carries the two baseline fields end to
end: schema columns on both session tables (SESSION_QUERY_SQLITE_SCHEMA
_VERSION 6 — derived, rebuilds in place), inserts, header selects, the
candidates CTE, rowHeader, sameHeader, and the cross-source
assertSessionHeadersCompatible — so a search hit's header keeps the
child's inherited confinement and conflicting live/persisted baselines
reject.
Red-first: out-of-range seedLength tests in both policy suites;
baseline round-trip and live/persisted baseline-conflict tests in the
session-query sqlite suite.
Review fix (ds-review-bot on #623): the unconditional slice regressed the
public SessionStore.fork path — a generic fork child gets seedLength but no
policy baseline, so slicing discarded its seed-carried sandbox/approval
switches and silently widened it to the deployment defaults (a forked
read-only/never parent produced a workspace-write/ask child).
overrideOf now branches on baseline presence: with a header baseline (a
delegation child) the fold covers only own post-seed switches — the
baseline captured from the parent's FULL log subsumes seed history; without
one, the whole log — seeded switches ARE the replayed inherited truth. The
permission preset fold scopes the same way. Red-first: generic-fork
seed-carried override tests in both policy suites.
The dedup guard pinned 'no duplicate stamp' — under the header baseline no
stamping exists, so its event-count assertion became a tautology and its
fence assertions a subset of the stale-seed case. The spec docblock claimed
the session invariants pin the stamping design (superseded); it now names
the three review-found timing threats the suite pins as distinct shapes.
Service overrideOf JSDoc shrinks to a pointer at the owning pure export.
The snapshot-coverage consequence names the deployment-strength limit, why
the automation protocol cannot express a parent-only override, and the
headless semantic-checkpoint precedent (seeded parent log + resume fixture
plugin) as the pickup shape — so the deferred scenario is not re-derived.
Review fixes (ds-review-bot warnings on #623):
- One chain, every consumer: the override resolution (own post-seed
switches ?? header baseline, closed-vocabulary validated) moves into pure
exports (sandboxOverrideOf / approvalOverrideOf); the services delegate,
and the permission presets consume them — current(session) and set()
now see inherited baselines, so a child inheriting danger-full-access
gets REAL knob switches when workspace-write is selected instead of a
silent no-op, and a seed-carried preset selection is subsumed by the
baseline. current(events) becomes current(session) (pre-release; the
only callers were tests).
- Unconditional durable validation: a malformed header baseline fails
loud on every read, no longer shadowed by an own switch.
- The two policy peers are declared optional (peerDependenciesMeta), so a
thin spawn/fork deployment without policy plugins can consume the
driver; verify-runtime-closure honors the flag.
Red-first: inherited-preset derive/switch-away and seeded-selection tests
in the permission suite; malformed-baseline-with-own-switch tests in both
policy suites.
CI per-file coverage flagged the new sandboxMode/approvalPolicy branches:
invalid-type cases in both session header validation tables, and a shared
persistence-contract case pinning that both backends round-trip the
baselines verbatim and keep absent fields ABSENT (presence is the signal
the policy owners branch on).
The cordis_inspect api surface embeds the SessionHeader declaration, which
gained sandboxMode/approvalPolicy; keyless refresh, model transcript
untouched.
Review fix (ds-review-bot critical #2 on #623): the first-turn event stamp
had a durability hole no turn anchoring can close — an idle SessionStart-
style injection persists a complete one-shot turn before any prompt turn
opens, so a crash in that window left a resumable-looking child with no
inherited policy, falling back to a possibly wider deployment default.
The captured overrides now ride the child's creation meta into its
immutable SessionHeader (sandboxMode/approvalPolicy, neutral strings at the
session boundary — the delegationDepth precedent), durable from the moment
the session exists: no listener ordering can starve the baseline and no
crash window can lose it. overrideOf(session) on both policy services
resolves fold(events past header.seedLength) ?? header baseline, validating
against the closed vocabulary on read; stampOverride and the prompt-submit
listener machinery are deleted. The header field rides both persistence
backends (JSONL header line; SQLite sessions columns, SCHEMA_VERSION 11 —
pre-release, no migration). pty-local reads through overrideOf so PTY
spawns see the baseline too.
Red-first: header-durability-before-any-turn test (the injection crash
window shape), baseline/seed-boundary/closed-vocabulary contract tests in
both service suites; the real-wall suite (race, veto, fork stale-seed,
grandchild) re-anchored on header assertions and green. The Agent Note's
Alternatives now records the superseded event-stamping iteration with the
review evidence; bilingual docs updated.
Three review rounds each found a new silent-loss path in the same
wrapper state (dispose racing an in-flight flush, overlapping hints
displacing the retained promise, the provider's fixed 30s flush timeout
rejecting while the processor still drains). Every path exists only
because forwarding the seam's turn-boundary hint to forceFlush() made
this backend the process's second flusher against undocumented SDK
internals from the upstream experimental tree.
The backend now implements no flush(): the batch processor is the only
flusher, its scheduledDelayMillis (already deployment-tunable through
the processor passthrough) governs export cadence, and shutdown()'s
drain is complete by construction. The two race-pin tests collapse into
one dispose-during-in-flight-batch drain pin; the seam's optional
flush() contract now tells implementers they own the concurrent-flush/
shutdown interaction. Removal rationale and the reinstatement trigger
(a stated turn-boundary latency requirement scheduledDelayMillis cannot
meet — and then via the processor's own forceFlush(), never the
provider's timeout-wrapped one) are recorded in the revival Agent Note,
both languages.
Resolutions: regenerate docs/capability-seams.md (the only conflict);
master widened the bilingual pairing scope to package READMEs, so this
merge adds the Chinese counterparts and i18n records for the three
telemetry READMEs and patches packages/README.zh.md with the telemetry
group row.
The automation-only ACP refactor (#586) removed the bridge's pendingSwitches
idle-preset anchoring the inheritance note cited as precedent. Replace the
dangling analogy with the load-bearing fact a reviewer actually needs: an
injection-triggered first turn (a SessionStart hook racing the prompt)
dispatches no prompt-submit and no model request, so the one-shot stamp
listener still lands before the child's first model request.
Review finding (Codex P1), pinned red-first: the marker was tied to
telemetry-plugin lifetime, but receivers key crash detection on its
absence per session. A normally closed session in a long-running host
retired silently (classified as a crash once stale), while a telemetry
reload marked every still-live session as cleanly ended.
The session/disposed handler now emits the marker at the session's own
termination edge before retiring it; the dispose-time sweep only marks
sessions still alive at application teardown (their own edge would fire
unobserved). READMEs restate the marker semantics: telemetry stopped
observing cleanly — a marker followed by more session events is a
telemetry reload, not a session restart.
Resolutions: keep both sides where master's new storage/workspace group
registrations sit adjacent to the telemetry ones (gen-doc-graphs
SERVICE_ROLES/GROUP_ORDER, tsconfig.base.json wildcards,
packages/README.md rows); regenerate the graph docs; condense four
packages/README.md rows back under the 790-word ceiling.
Post-merge reconciliation: master's plan-mode command wording and tool-schema
changes land in the recorded scenario's keyless sidecars (refresh, not
re-record — the model transcript is untouched); the abort-handoff fake agent
stubs ctx.get for the driver's new synchronous inheritance capture; the two
zh Agent Notes gain the sentences their English sides added on this branch,
re-recorded in the pairing manifests.
Review fixes (ds-review-bot on #623):
- Capture-at-delegation: the driver now reads overrideOf(parent.session)
for both knobs synchronously before its first await, and the prompt-submit
listener stamps those captured values — a parent switch racing the child's
asynchronous creation belongs to the parent's future, not the child. The
inheritOverride(parent, child) service method is split into its two halves
(overrideOf / stampOverride) accordingly.
- Veto safety: the one-shot prompt-submit listener registers with
prepend: true, so a veto-capable listener (a denying UserPromptSubmit
hook) cannot close the child's first turn without the durable stamp.
Both regressions are pinned red-first in inheritance.spec.ts: the
delegation-vs-late-switch race (delegate tool flips the caller wider while
the creation transaction is pending) and a blocking prompt-submit listener
(stamp survives a promptless first turn). Service contract tests renamed to
the split API; READMEs and the bilingual Agent Note updated.
Per-session policy overrides (sandbox/mode, approval/policy) never crossed
the delegation boundary: a spawn child of a read-only-switched parent ran
under the wider deployment default, and a fork child missed any switch made
after its seed boundary — delegation was a bypass channel for a user's
tightening.
The in-process driver now snapshots the delegating parent's override chain
and stamps it onto the child through the canonical write paths
(SandboxPolicyService.inheritOverride / ApprovalService.inheritOverride),
anchored inside the child's first turn via a one-shot agent/prompt-submit
listener: turn-enclosed (durable), ahead of the first request (an inherited
'never' reaches the child's first system prompt), and positioned after any
stale fork-seed switch so the ordinary last-event-wins fold resolves it.
Only overrides are copied — an unswitched parent stamps nothing and the
child follows the live deployment default; both services are consumed
opportunistically, so compositions without them delegate unchanged. Nesting
composes by construction (each stamp folds the already-stamped parent log).
Evidence: inheritance.spec.ts drives scripted-model children into the real
dsh-fs-sandbox fence through the real write tool (disk-state + denial-marker
assertions; spawn, stale-seed fork, grandchild, escalation fail-closed, and
no-stamp guards), inheritOverride contract tests in both service suites, and
the recorded subagent-sandbox-inheritance ACP snapshot (read-only preset →
delegate → child denied, replayed keylessly).
See .agents/notes/implemented/feature/2026-07-25-subagent-policy-inheritance.md.
Second review round, both pinned red-first:
- Overlapping turn-boundary flush hints now JOIN the outstanding flush
promise (Promise.all) instead of displacing it: the SDK's
concurrent-flush guard resolves an overlapping forceFlush()
immediately, so retaining only the latest promise let shutdown()
proceed while the first export was still in flight — the same silent
drop the single-flush fix closed.
- Adoption replay contains failures per event, matching the firehose:
one rejected record is withheld fail-closed while the rest of the
historical log still hands off. Wrapping the whole loop let a single
failure silently skip the remainder on an already-adopted session.
Three review findings, each pinned by a red test first:
- The OTel backend retains the latest turn-boundary flush promise and
shutdown() awaits it before provider.shutdown(): the SDK's
concurrent-flush guard makes the shutdown-internal flush return early
while one is in flight, silently dropping everything enqueued after
the flush snapshot (including the coordinator's dispose-time shutdown
markers).
- The coordinator retires sessions from the adopted set on
session/disposed (mirroring the persistence coordinator): a long-lived
backend no longer retains closed sessions and their event logs, and
final unload no longer stamps shutdown markers for dead sessions.
- The exporter config passes through whole to OTLPLogExporter and its
type/JSDoc now advertise the full OTLPExporterNodeConfigBase shape:
rebuilding url/headers only silently dropped documented SDK options
(timeoutMillis, compression, keepAlive, ...).
tests/otel.e2e.ts self-skipped on $DSH_OTLP_E2E_ENDPOINT, which nothing in
the repo sets — it never ran; the mock-collector wire spec and the keyless
Loader-composition e2e already cover the pipeline both ways.
The severityOf compact/end probe parsed another package's merged event
shape by string comparison — an untyped cross-package contract that breaks
silently — and its only consumer was the test's own stand-in declaration.
Unknown event types now uniformly fall through as info; outcome semantics
stay with the owning package.
The seam keeps the telemetry/redact scrubbing interface but ships no rules
of its own: the innermost next() passes records through unchanged, and
deployments mount their rules as waterfall listeners. As an SDK we cannot
know which patterns are secrets in a given deployment; a shipped list
invites false confidence while catching only known shapes, and false
positives would corrupt exported bodies. Mechanism stays with the seam,
policy moves to the deployment; both READMEs and the Agent Note state the
raw-export default plainly.
The loader-composition e2e now mounts a deployment-style rule fixture and
pins the same wire behavior: secret absent, placeholder present, canonical
log untouched.
Revive the reviewed session-telemetry packages from the closed
session-telemetry-otlp-rfc branch (PR #222/#231) on current master, renamed
to @deepseek-ai/dsh-session-telemetry{,-otel} (the SDK component-telemetry
package holds the dsh-telemetry name).
Delta over the branch version: every record now passes a telemetry/redact
waterfall between projection and emit() — the innermost next() applies a
non-configurable conservative credential-shape rule set, listeners stack
stricter rules, a throwing rule withholds the record fail-closed, and the
canonical log is never rewritten. This answers the export-side concern that
closed PR #222; the boundary axiom (our aspect ends at emit(); delivery is
the reporting SDK's) is unchanged, and the runtime-telemetry RFC's outbox /
readCommitted lane is recorded as deferred in the Agent Note.
Covered by seam/redact/OTel-wire unit tiers (100% per-file) and a keyless
Loader-composition e2e that boots the examples fixture against a mock OTLP
collector and pins redaction on the wire plus the untouched canonical log.
17 master merges (GUI host/web stack, compact header, dshweb, fast local
hooks, windows coverage skips, worktree runtime). Conflicts:
- packages/ui/user-interaction/src/index.ts: master extracted the question
types to types.ts; took the re-export and grafted our detail field (the
plan-review payload) into the extracted AskUserQuestionItem.
- generated catalogs regenerated over both sides.
The module graph crossed mermaid's default 500-edge render guard with the
GUI packages plus our plan-mode edges; raised maxEdges in verify-mermaid
(a secure config settable only at initialize).
Note: the remote branch also renamed packages/mode -> packages/plan
(PR #512 plan-mode-simplification, fast-forwarded before this merge).
Conflict: the generated event-producer-consumer doc — regenerated over merged
sources. Also strips the trailing whitespace / extra EOF blank line that
master's vitest.config.ts and client api.ts arrived with (the pre-commit
whitespace hook rejects any commit touching them otherwise).
Wenlu Wang