Overlaps master on the preset e2e, the web scaffold, and the generated
config-catalog triplet; all merged textually, so the catalog is regenerated and
its pairing hashes re-recorded rather than trusted.
wireSchemas() already advertised only run_code under mode: 'code', but the
executor resolved every call through get(), which returns the full visible
map plus the reserved transport. A model could name a native tool directly
and bypass run_code entirely. Route the execution-path lookups through a
new private resolveExecution() that applies the mode collapse at the
operation boundary: model-direct calls under 'code' may only name run_code
(UNKNOWN_TOOL otherwise), while SDK sub-dispatches (parent token set) keep
every visible tool. get()/schemas() public semantics are unchanged.
The denial happens at createExecution, before the extensible policy
pipeline — pre-execute listeners, approval ask, and guards never observe
a call that is deterministically denied. A collapsed call honors the
pre-dispatch cancellation contract, routes aborted results through the
visible tool's finalizeContent, and captures the finalizer before
argument materialization.
Regenerated docs, catalogs, graphs, scoped events, re-recorded
translation pairs, and updated test assertions.
Fixes#1815
`Config.roots` defaulted to `[]`, so a deployment that did not patch both roots
in got a roster with none — `dsh run` once booted exactly that and failed
resolving `standard`. The user root never needed an app: `<dshHome>/.agent-presets`
is the same place in every deployment, resolvable here the way
`dsh-skill-local` resolves `<dshHome>/skills`.
The roster now derives that root itself unless `includeUserRoot` is false, and
`apps/cli` supplies only the SHIPPED root, whose path an installed app alone
can resolve. The derived root is appended after every configured root, so a
shipped id still shadows a home directory claiming it and `writableRoot()`
still prefers an explicitly configured `user` root; the set is resolved once at
construction, because a root set that changed between a `list()` and the
`copy()` acting on its answer would author into a directory the caller never
saw.
Every test that pins an exact roster now says `includeUserRoot: false` — the
machine's real harness home must not decide what a golden or an assertion
counts.
A continuable child that stopped without reporting — an error, a token
ceiling, cancellation, teardown — left its parent nothing to act on.
The continuation manager now delivers an unconditional settlement
notice to the durable direct parent before releasing ownership, folding
consumed work (foldConsumedWork supersedes findLastMessageTurnEnd) so a
claimed-but-unrun prompt reads as aborted rather than completed, waking
an idle parent, steering a busy one, and never waking a closing tree.
1504 hand-written ranges pointing at workspace members become workspace:^, so
pnpm pack substitutes each member's real version at publication: sibling
peerDependencies follow the family version instead of being pinned at ^0.0.1,
and a reference to a vendored package follows that package's own line. Without
this, publishing 0.0.2 ships peer ranges naming a version that does not exist,
and 0.0.1-rc.1 does not satisfy ^0.0.1 either.
It also retires ranges that had gone stale against the workspace: ^4.0.0-rc.6
for a 4.0.0-rc.7 checkout, ^3.17.0 for schemastery 3.18.0.
workspace:* stays where an exact published version is the point, which is how
the Landlock entry pins its platform packages.
A workspace constraint now requires the protocol, so a new package cannot
reintroduce a hand-written range. The same constraint caught packages/boot/cmdline
arriving on master without the publishable trio, which this change completes.
Every package under packages/, apps/, and vendor/ drops "private": true and
declares publishConfig.access "restricted": the repository now states which
packages it publishes instead of deciding it at publish time. Each one also
declares its repository and directory, which is how a consumer of a private
package reaches its source.
The Landlock packages move to restricted with them. They have never been
published, so nothing anonymous depends on them today, and the whole
@deepseek-ai scope stays private.
The workspace constraint that required every package to be private now applies
to non-members only, and asserts the publishable trio on each release member.
Machine-produced by `pnpm run rescope-vendor --apply` plus the regeneration it
prints: `pnpm install` for the lockfile, `pnpm run gen-third-party-notices`,
`verify-translation-pairing --write` for the touched bilingual pairs,
`gen-doc-graphs`, and one typert snapshot whose ids embed character offsets.
`pnpm run rescope-vendor --check` verifies the result.
Renames nine vendored packages (cordis, cosmokit, schemastery and the six
@cordisjs plugins) and every reference that resolves them: manifest names and
dependency keys, module specifiers including declare-module merges, cordis.yml
plugin names, tsconfig paths, every Markdown fence, and `docs/` prose.
Directory names, upstream versions, and dependency ranges are unchanged, so
vendor/README.md still reads as an upstream snapshot; its manifest table gains
an upstream-name column so THIRD_PARTY_NOTICES keeps MIT attribution pointed
at each fork's origin.
The tutorial tier follows the rename end to end: its yaml fences named plugins
the Loader can no longer resolve, its `ts ignore-check` fences disagreed with
the compiled fences beside them, and its prose quoted both. The contracts that
told readers to keep upstream names — the root convention and the vendoring
cookbook's tree comment and manifest invariant — now say to rescope instead.
Two rules read `@deepseek-ai/` as "another workspace plugin": the client bundle
purity gate now names the vendored libraries a browser bundle inlines, and the
files where a bare `cordis` is an agent-preset id keep that product data.
A restriction was compiled against the global tool layer alone: only
global-layer tools were tested against `admits()`, and every chain-layer
tool was overlaid unfiltered afterward. That read the exempt set as "the
global layer" when what it means is "what this scope registers itself" —
two descriptions of the same set only while every model-facing tool sat in
the host composition.
Moving those rows onto the agent plane separated them. A preset's tools are
an ANCESTOR contribution to a joined agent, so a subagent's `toolFilter`
stopped constraining anything it was given; and with the global layer empty
`restrict()` rejected every name it received as unknown, failing the child
outright. With the same tools in the global layer the filter still admits
and applies normally, which is what makes this a regression of the move
rather than a standing limitation.
`view()` now filters everything a scope inherits — the global layer and
every ancestor layer on its chain — and exempts only the layer the scope
owns. That exemption is load-bearing rather than incidental: the delegation
runtime registers a child's `report` and structured-output tools into the
child's own layer, and a filter naming the capabilities the child may use
must not strip the machinery it answers through. Tool order, and with it
prefix-cache reuse, is unchanged: inherited names keep their global-then-
ancestor position and own-layer names still come last.
The diagnostic said "unknown global tool" while listing what is really the
inherited surface; it now names the surface it checks and says why an
own-layer name is not restrictable.
Fixes#2185
`session.create` also adopts an already-live session, and the preceding
commit newly allows adopting one under the preset it switched to while
blank. Its response still echoed `header.agentPreset`, so that adoption
answered with the preset the session had just left — contradicting the
request it had accepted and the row `session.list` serves for the same
session from `resolveSessionPreset()`. The echo now resolves the same way.
The `assertPresetUnchanged` parameter doc said `existing` was the preset
the session was created under; both callers now pass what it runs.
`composeFrom()` was documented as "infallible" and "cannot fail" beside two
`@throws`. It has no composition failure mode — no roster read, no mount, no
file — but it does reject a caller error, and the wording now says which.
The package-level "switched preset" test re-linked to the same preset id,
so it could not tell reading the parent's live scope chain from reading its
creation header. A second fixture preset makes the switch real.
The Web browser lane's subagent goldens gain the preset badge a child now
shows, which is the visible consequence of recording its composition. That
lane runs only under DSH_EXAMPLE_MODE=lib and was missed before.
The Agent Note records two limits found in review: a cold-resumed
continuable child joins its parent's current composition rather than the one
its header names, and `toolFilter` does not constrain a joined child. The
latter is a regression from the agent-plane move rather than anything this
change introduces — with the same tools in the global layer the filter
applies normally — and is tracked in #2185.
Refs #2185
A delegated in-process child now acts only within the sandbox scope fixed
at delegation: captureDelegatedPolicyOverrides still snapshots the parent's
explicit sandbox override but pins the child approval policy to 'never'
(instead of inheriting the parent's), so every child ask — sandbox_permissions
escalations included — is rejected deterministically by ApprovalService
before any answerer, with the audit pair still logged. Every in-process
child additionally receives the scoped subagent:delegation runtime-context
statement telling it to report a scope limitation instead of retrying.
Supersedes the approval half of the policy-inheritance decision (new Agent
Note cross-linked from both prior notes and the approval-seam Q&A); refreshed
child snapshot fixtures carry the pinned event, and
subagent-published-run-failure now persists a one-event child log.
Tool and prompt-section visibility is inherited along dsh-scope's parent
chain, and an agent's scope key is minted with no parent. Per-session agent
presets moved every model-facing row onto the agent plane and made
AgentPresets.mount() the one thing that binds that link, from the api-proxy's
session create, resume, and fork paths. The two in-process subagent drivers
installed only the per-child persona and tool filter, so a child's scope chain
had length one and its registry view resolved the global layer alone — which
is empty wherever a preset roster is composed. One-shot children reached the
model with no tools, continuable ones with only the host-plane `report`, and
neither carried its parent's persona, workspace context, or skill catalog.
AgentPresets.composeFrom() joins one agent to the standing composition another
already runs on. It is a bind, not a mount: the child gets its parent's exact
generation, so a composition edited since the parent started cannot fork it
onto another one, and it is synchronous, which is what lets a child creation
window use it. applyChildComposition() now takes the parent and performs the
join first, making a child composed without it unrepresentable at the call
sites. childSessionMeta() records the joined id so a cold read rebuilds the
composition the child actually ran under.
The audit that followed found two api-proxy readers on the wrong authority:
presenterScopeFor() and the live-agent branch of assertPresetUnchanged() both
read header.agentPreset, which goes stale the moment a blank session switches
preset. A switched session's cold transcript resolved presenters in the older
composition's layer and silently degraded to generic cards, and the gateway
refused to adopt a live session under the preset it actually runs while
accepting the one it left. Both now resolve through resolveSessionPreset(),
matching the resume branch fifteen lines above. The owning architecture Agent
Note carried the stale claim that the header records what a session runs; it
is corrected to name the header/log pair and its three readers.
Fixes#2165
Address ds-review-bot on #2127:
- assistant-output: the rule has ONE implementation, the incremental
AssistantOutputFold (push/pushText/collect); finalAssistantOutput folds a
complete suffix, the SDK backend folds notification events, and the ACP
backend folds raw chunk text into the same streamed fallback.
- subagent/end.lastAssistantMessage: 'no output' is encoded once — absent,
never [], on both lifecycle shapes (observeRun now omits empty output).
- tool-subagent: a non-completed foreground result stays isError but appends
the child's preserved partial text after the stop-reason headline.
- Authored keyless snapshot scenario subagent-max-tokens-partial pins the
assembled transcript: the child's committed log carries the usage-only
empty message and the parent's tool result carries the partial answer.
- Rule-boundary sentence (message wins over later streamed text) and the
consumer half recorded in the Agent Note; comments trimmed to pointers.
- child-agent.ts declares both policy-service augmentations as explicit
empty type imports, so removing the ApprovalPolicy import cannot
silently degrade ctx.get('approval') typing.
- dsh-subagent-inprocess no longer consumes the policy services in src,
so its optional peers and tsconfig references are dropped; both
policy-inheritance Agent Notes state the current ownership.
- The continuable suite pins that a later child-side switch beats the
delegation snapshot and that an unswitched fork parent seeds no
policy events.
A continuable background child (the default backgroundMode for both
delegation tools) never received the parent session's explicit
sandbox/approval overrides: materialization applied only child
composition, so a danger-full-access parent produced workspace-write
children whose every out-of-workspace operation raised an approval
prompt.
Move the one-shot driver's capture/append pair into the shared
child-agent module (captureDelegatedPolicyOverrides /
appendDelegatedPolicyOverrides) and call it from both paths:
startContinuable captures before its first await, only fresh
materialization appends the source-tagged events (after any fork seed),
and a cold resume replays the persisted delegation events instead of
re-capturing the parent.
Adds the continuable inheritance unit suite, the ACP snapshot scenario
subagent-continuable-inheritance (fails without the fix), the
continuable policy-inheritance Agent Note, and the seam-level README
contract, with bilingual counterparts.
Fixes#1692
A max-tokens step that assembled only tool-call blocks appends an
EMPTY-content assistant/message (the usage host). Three consumers each
hand-rolled output selection and all let it erase the child's real
answer: the in-process readResult and the Activation subagent/end
capture took the last message unfiltered, and the SDK backend let any
message beat its streamed-text fallback; the in-process driver also had
no streamed-text fallback for cancelled turns.
dsh-subagent now owns the canonical rule in src/assistant-output.ts
(last non-empty assistant message, else the accumulated text-delta
stream) and all three consumers apply it. Regression tests in all three
packages fail under the previous selections.
Closes#1514
git mv per the regrouping RFC: the five human-collaboration seams and
tui join packages/interaction/, app-boot becomes packages/boot/, and
jsonrpc joins the renamed scaffold/ (formerly sdk/) as its server half
beside client/protocol/create-sdk/helper/scripts/telemetry, whose
folders drop the legacy sdk- prefix. Three new group README triplets
replace the ui/ and sdk/ ones; tsconfig references/paths/globs,
knip keys, vitest globs, gate scripts, catalogs, docs, and the
lockfile follow. Adds the four settled FIXME rename markers
(dsh-sdk-server, dsh-sdk-telemetry, dsh-sdk-helper, dsh-sdk-scripts).
The scaffold folders diverge from their npm names until those renames
land, so tsconfig.base.json maps the three affected names explicitly
beside the group wildcard. Also repairs two pre-existing stale-path
classes the strengthened sweep surfaced: docs/web-styling.md's retired
web-ui host package and type-model spec fixture-literal joins.
app-boot's three Loader-composition specs time out at the default 5s
under full-suite parallel load on this filesystem (pre-existing;
pass isolated with --testTimeout=30000); interaction/scaffold/boot
suites otherwise green (687 passed).
All agent/* and agent-loop/config-start-failed events take one payload
object carrying the agent subject; waterfall/serial payloads require a
signal and keep next as the final argument. PreStepContext and
RequestFailureContext are unfolded into payloads and retired.
goal/changed follows the same shape so agentEvents keeps its listener
error containment. ReactLoopAgent builds its scope carrier once in the
constructor. Regenerates scope resolvers, tool-cordis api catalog, and
docs catalogs; updates all affected listeners, tests, and the
core-data-structures docs (en + zh).
Rewrites the service API table, authority-versus-provenance contract, residency
routing, and deferred-work list; scopes the in-process driver README to one-shot
runs; and restates both model-facing tools' outputs, which no longer carry a
task id.
Replace the Task-backed continuation manager with one durable Session plus at
most one process-local Activation — a residency epoch for a reconstructed child
Agent, not a request, result, cancellation, or Task boundary. The manager owns
activation admission, authority, the live ownership graph, cold resume, and
child-first disposal; the Agent inbox is the only turn FIFO.
- startContinuable() is async and returns { childId, messageId } at inbox
acceptance; followup() takes a SubagentAuthority and returns AgentMessageId.
- SubagentProvider.resume?(), SubagentProviderResumeRequest, SubagentRun.steer?(),
SubagentProviderStartRequest and SubagentContinuation are deleted;
prepareContinuable?() is the continuable-creation capability.
- Cold resume calls ctx.agents.resume() from the manager through a private
activation-owner scope, never dispatching through a provider.
- Extract shared child composition, descriptor seeding, depth accounting, and
one-shot run settlement so the manager and one-shot driver keep one home
per fact.
Tests and docs follow in subsequent commits.
