New note records the dsh web composition decision (flat cordis.yml, the
AppCLIEntry/AppWebEntry class pair, one declaration place per config
source, the five-way transport split) with its rejected alternatives; the
client-plugin-loading note's roster endgame and HMR sections are brought
current with what shipped.
apps/cli/cordis.yml holds the whole composition flat — the host runtime
rows, the api-gateway row, the webserver row, and the ten dshClient rows.
AppCLIEntry is the pre-cordis glue: layered env (ambient > cwd .env >
$DSH_HOME/.env, fixing DSH_HOME=... dsh web not finding its key), patch
composition from the three non-yml sources (profile json through the static
PROFILE_MAPPINGS table, CLI flags, the resolved frontend distIndex), the
Loader include boot (--dev appends the hmr row before the settle), and the
fail-loud triple (assertEntriesLoaded + installFailLoud + an all-ACTIVE
sweep for PENDING fibers). web.ts shrinks to argv parsing + the URL line.
The four free functions in boot.tsx become one kernel class holding what
must exist before cordis: the parsed BootManifest, the ClientModuleSystem
instance, and the loading-page handles. Context/Loader setup runs in
parallel with the immediately-tier prefetch, but entry creation awaits the
prefetch: materialization is tree.import's synchronous require, so
cross-package require edges (i18n -> runtime/client) need every
immediately-tier factory registered first — unbarriered creation raced
10-25% of boots. The kernel adopts the modules entry (writes the
__DSH_MODULES__ slot pre-cordis, creates the entry first, skips its graph
row), and provide('modules') now lives in the adoption apply. apps/web
drops its host-package edges (composition is apps/cli's job).
connection binds the web transport: it injects httpServer + apiProxy and
registers toFetchHandler(ctx.apiProxy) under the /api prefix (the node:http
to fetch bridge moves in from the webserver, keeping the res-close disconnect
detection and drain/close backpressure waits). hmr owns dev reload: a
stat-poll watch per graph row driven by clientModuleHost.onGraphChanged,
rebuilt(id) on content change, and the /plugins/events SSE route (GET/HEAD
guarded); frame types are single-sourced in events.ts shared by both halves.
The node half is ClientModuleHostService (ctx.clientModuleHost): it composes
the __DSH_BOOT__ graph by scanning loader entries for dshClient packages,
serves /plugins/<id>/client.js, taps the index render, and exposes
rebuilt/onRebuilt/onGraphChanged. Scanning is incremental per package — no
full-rescan path exists: internal/plugin marks the fiber's entry name dirty,
a flush reconciles each name against live entries, package metadata
(including negative verdicts) caches forever, and re-hashing is reachable
only through rebuilt(id). The browser half moves wholesale to the standard
./client export (ClientModuleSystem, parseBootManifest with the dual-view
BootManifest, and the adoption plugin face that reads the
window.__DSH_MODULES__ slot and provides ctx.modules).
HttpServerService provides ctx.httpServer: register(route) -> disposer
(duplicate patterns throw), tapIndex transforms in registration order, and
the bound port; matching is exact > longest prefix > static dist fallback
(403/405/SPA semantics preserved). The server listens on activation, answers
per-request failures with 400 + a log line instead of exiting the process,
and knows no harness concepts — the boot graph, bundle routes, SSE channel,
and /api prefix all moved to their owning plugins.
createApiProxy moves from dsh-host-runtime into dsh-host-apiproxy (the
dependency direction already pointed this way); the package now
default-exports ApiProxyService (config {provider, model}, provides
ctx.apiProxy) while staying transport-agnostic — it registers no routes.
Runtime keeps bootHost/startHost for the headless path with its import
re-anchored, and drops the mountWebPlugins roster mounting helper.
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, ...).
Master's i18n batches added Chinese counterparts to ~50 docs this PR
edits in English. Bring each zh side along with the minimal edits
covering the en diff (recorded-hash diffs, not re-translations),
reunite the stream-workflow-progress pair under rejected/ with its
manifest entry, re-record all pairing hashes, and regenerate the
event/persistence/tool catalogs and doc graphs over the merged tree.
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.
The dev bundle watch missed rebuilds that landed while the registry was
constructing: fs.watchFile captures its comparison baseline with an
ASYNCHRONOUS first stat, so a write racing that window is absorbed into
the baseline and never reported. Standalone repro missed 24/400 same-tick
rewrites; the CI flake in web-plugins.spec.ts ('watch mode: a bundle
content change re-hashes the row...') was exactly this — the spec writes
immediately after createHostWebPluginRegistry returns.
The watch now polls from one registry-owned setInterval against a stat
baseline the scan itself captures synchronously, stat-before-read: a
write landing between stat and read leaves the hash newer than the
baseline (next tick re-hashes to the same rev, no spurious notify); a
write landing after the read leaves the baseline older (next tick
detects and notifies). No blind window. The poll iterates the live
table, so rescans retarget the watch for free and dispose clears one
timer. Stress: real-registry same-tick rewrite 0/600 missed (was 1/300);
spec watch tests 0/50.
New regression test pins the same-tick-as-construction write. Its
rewrite deliberately differs in size from the seed: a same-millisecond
same-size rewrite is invisible to any mtime+size poll (coarse fs
timestamps) — a stat-polling limit, not this regression. Loading-model
Agent Note updated in both languages (pair re-recorded).
The pool selector is defense-in-depth only — pull_request executes the
PR's own workflow definition, so YAML cannot enforce runner trust. Make
the actual enforcement boundary explicit in the decision record:
org-side disabled forking (the public release is an isolated read-only
mirror under a separate org), with migration to a repo-restricted
org-level runner group with base-branch workflow pinning as a hard
gate before forking could ever be enabled.
The hosted 32-core runner is exclusive to one job, but the vm-backup
pool shares one 64-core VM across four runner instances; concurrent
PRs could stack 4×24 = 96 Vitest workers and re-trigger the documented
aggregate-contention failures in the timing-sensitive process suites.
Bound the self-hosted leg at 12 workers per job (48 host-wide fully
loaded) and keep 24 on the hosted leg, selected by the same expression
as the pool.
The row rendered a status only for error, so a call cancelled before
tool/result read as a completed plan update even though no todo/write
occurred. Non-ok states now ride the generic row's StateDot semantics
(ongoing dot while running, warning dot + 已中断 marker when interrupted);
the ok badge stays for settled successful updates.
Keep inherited child prompt markers bounded by the normal silence fallback. Stage web-plugin rescans atomically and retain missing watch state until a successful rebuild.
additionalProperties: false on the todo_write item schema is model-visible
(tool schemas ride the request header and the code-mode prompt types), so
the pinned ACP/headless expected outputs re-record. Keyless refresh; the
two locally-failing scenarios are this machine's known environment issues
(HOME-symlink cwd normalization, SQLite ExperimentalWarning), not the diff.
Sweep all remaining sources that still described coverage as an
enterprise 32-core job: the ci.yml jobs preamble, the three-job
paragraph of the larger-hosted-runners note, and the required-pool
sentence of the portable-recovery note — English and Chinese sides of
both notes, with their i18n pairing records re-recorded.
Keep the cache restore for the ephemeral hosted (untrusted-PR) leg where
it is a genuine speedup, gated by the same expression as the runs-on
pool selector; the self-hosted leg skips it and installs from the
persistent local store.
master gave 2026-06-29-todo-write-tool a Chinese counterpart; the English
side's web-consumer sentences now translate across (in-body links keep
their .md targets per the pairing contract) and the pair is re-recorded.
The todo-panel fake gains the time/callTime fields ToolResultNode now
requires.
- Route untrusted PRs (forks + Dependabot, same author test as e2e.yml)
back to the hosted enterprise pool via a runs-on expression: Dependabot
PRs are same-repo, so the previous head.repo guard admitted
dependency-supplied code onto the persistent self-hosted VM. A single
job with pool selection keeps all-checks-passed free of skips.
- Drop the pnpm-store cache restore from this lane: on self-hosted the
hosted-path cache actually HIT (Linux key) and spent ~52 s pulling
181 MB into a path pnpm never reads; the persistent local store
already serves warm installs in seconds.
- Update the larger-hosted-runners Agent Note (en/zh + i18n pairing
record) so the decision record describes the shipped topology:
coverage on the in-house vm-backup pool for trusted PRs, hosted
Ubuntu 24.04 32-core retained for untrusted PRs.
macOS symlinks /tmp → /private/tmp; when the session header cwd lacks
the /private prefix but fs tools resolve symlinks, tool result paths
get the prefix and the cwd replacement never matches. Strip the prefix
before cwd→{{cwd}} replacement so both forms normalize identically.
Also extend refreshFixtureReplacements to match spill paths by filename
suffix, preventing raw fixture churn when a refresh runs from a
different session directory.
The projection sentence prescribed resetting todos on window rebuild —
the implementation deliberately preserves the tail-page seed and lets
only in-window/live writes overwrite. The toolview section named a
nonexistent ctx.toolviews registry — the shipped seam is the keyed
conversation.chat.toolview slot via ctx.slots.register. Both sides of
the bilingual pair re-recorded.
additionalProperties stays true in the published schema while execute
rejected extra keys, so generated typings and validation disagreed with
runtime behavior. The item schema now declares additionalProperties:
false — the registry's arg validation rejects extra keys with a
path-qualified violation before execute runs — and the redundant manual
check is dropped (tool catalog regenerated).
repairGap installed the repulled window without the response projection;
a todo/write missed during the gap and already outside the new tail page
kept the stale list. The spec pins adoption through the repair path.
sessionHistoryValueSchema declared only events/hasMore, so the fetch
carrier's Zod parse stripped the tail page's todos projection — the
in-process and fixture paths carried it while a real WebApiClient lost it.
The fetch-carrier spec pins the field through the wire round trip.
The client's todos projection derived only from the paged display window,
so reopening a session whose last todo/write preceded the tail page showed
an empty plan until the user paged back — session-level state cannot be
reconstructed from an arbitrary window. The host owns the full log, so the
tail history response now attaches todos (latest todo/write backscan, the
same posture as the view pairing); installWindow seeds it, window rebuilds
preserve it, and any in-window or live write keeps overwriting it. The
fixture mirrors the host; docs and both Agent Notes record the mechanism.
runtime README documents ConversationSnapshot.todos and its window-scoped
limitation; the todo tool README and Agent Note name the web client among
the event consumers; the web display note records the cold-load gap and
fix directions (bilingual pair re-recorded).
Eight real bundles through the DI chain in ?fixture mode: plan strip
content, dedicated row summary + details linkage, collapse hint, zero page
errors — the CI-gated assembled-surface coverage for the todo display.
The tool appends the snapshot mid-execution, between tool/call and
tool/result; the fixture spliced it after step/end with a post-turn
timestamp, so acceptance never exercised the production ordering. A spec
pins call → snapshot → result with monotonic times.
null roots, non-object roots, and null array items (retained verbatim on a
rejected tool/call) now take the documented generic-summary fallback instead
of throwing into the row error boundary.
An item carrying keys beyond content/status (ids, children, priority) was
flattened to {content, status} on append, so the logged snapshot diverged
from what the model believed it wrote (model-visible must equal logged).
Reject loudly; the isError result lets the model self-correct.
Turtle
imccyu
Yichen Jiang
kingwl
Tianyi Cui
Chinesezjc
_Kerman