Review feedback (tianyicui, PR #657 inline): the missing-service message
should mention dsh-tasks, which defines ctx.tasks, rather than promoting a
specific backend. The seam's own surfaces (README, the direct-mount fence)
keep pointing at implementations, so the pointer chain still lands on
dsh-tasks-local without the producer strings going stale when another
backend becomes the recommended default. Agent Note updated accordingly
(en+zh, re-recorded).
The tasks/ family now matches the capability-seam shape: @deepseek-ai/dsh-tasks
keeps the abstract TaskService (ctx.tasks contract, vocabulary types, snapshot
invariant companion) and the new @deepseek-ai/dsh-tasks-local carries the
process-local registry (LocalTaskService: in-memory store, settlement,
owner-cleanup effects, teardown, TASK_WAIT_TIMEOUT). Compositions and test
harnesses now load dsh-tasks-local; producers, TaskKindMap merges, and
dsh-tool-tasks keep importing the seam only.
Producer misconfiguration diagnostics name dsh-tasks-local because loading the
implementation is the fix. The registry behavior suite moves to tasks-local;
the seam keeps a stub-subclass registration test and the probe-based invariant
suite.
An omitted maxDepth meant unbounded recursion, and the shipped examples
shipped that default. maxDepth now defaults to 1; a numeric cap requires
the provider's depthLimit capability (the mount fails loud and points to
the explicit 'provider-managed' opt-out for out-of-process providers),
and a child AT the cap loses the delegating tool from its own schema via
the child toolFilter — prompt-face hiding on top of the execution-face
depth check. Examples pin maxDepth explicitly. The ACP snapshot harness
gains Scenario.childToolOmissions so a child session may legitimately
omit declared delegation tools from its pinned header and prompt;
affected subagent/workflow goldens are re-recorded.
The background-task change repeated its lifecycle design across implemented RFCs, package READMEs, JSDoc, test commentary, and model-visible schemas. That repetition obscured the contracts that maintainers must preserve and added avoidable prompt tokens.
Rewrite the implemented RFCs around the current design, keep authorization, exact-owner cleanup, wait/abort ordering, producer quiescence, and teardown-failure guarantees at their owning surfaces, and remove peer surveys, review history, control-flow narration, and emphatic restatement.
Shorten the task and subagent schema wording, synchronize the bilingual tool cookbook, and regenerate the config, service, RFC, tool, and replay snapshot derivatives. Runtime behavior is unchanged; test edits update prose-only assertions and descriptions.
Rewrite the agent-scope RFC with executable examples and an explicit security non-goal. Harden subagent scalar and depth validation, and pin live tool-filter semantics across code, tests, and generated docs.
Drain idle injection flushes before agent teardown, snapshot approval and subagent provider inputs, and gate subagent lifecycle events on real child readiness. Align the RFCs and generated contracts with the hardened behavior.
enableRunInBackground: false removed the parameter from the advertised
schema only — the arg validator deliberately allows undeclared keys, so
a caller (or a model that has seen the parameter elsewhere) could still
force run_in_background: true and start background work past the
deployment's opt-out, in both tool-bash and tool-subagent. Both
producers now refuse the forced key loud in execute(); tests pin the
refusal (and that nothing spawns) alongside the untouched foreground
path; the schema-omission-is-advertising rule is recorded in the
runtime RFC and both READMEs.
start({ kind, label, owner, run }) preflights everything that can fail
(the attachSurface fence, validation, the owner-cleanup attach) BEFORE
invoking the producer's run() starter, then commits atomically —
'work started but never got a collectable id' is now structurally
impossible instead of a producer try/catch rollback obligation (the
P1 review fix, rebuilt on #185's declare/execute split). Producers
lose their catch-wraps; the leak tests now pin the stronger property
that a failed preflight never spawns anything. TaskRegistration splits
into TaskStart (identity + run) and TaskHooks (cancel/done/readOutput);
docs, type-equiv manifest, catalogs, and both RFCs move with it.
One shared ctx.tasks registry (branded <kind>-N ids, owner-fenced
read/kill/wait/list, attachSurface misconfiguration fence, reported-flag
notice dedup, atomic register) + dsh-tool-tasks (task_output/task_list/
task_kill, completion-notice injection, background prompt habit).
Producers opt in via their own enableRunInBackground config: bash
(stream kind; seam slimmed to resolve/run/start returning a BashProcess
handle, bash_output/bash_kill deleted) and subagent (final-output kind;
done settles after run.dispose()). Owner disposal drains tasks through
the new awaited ctx.agents.onCleanup seam in the loop's disposal chain.
Both RFCs moved to implemented/; docs, catalogs, snapshots re-pinned.
The partial-toolFilter materialization fix (da6c6d58) stopped one field
short: the adjacent agentOptions key in the SAME Config has the same
schemastery trap. An omitted agentOptions materializes {}, which is truthy —
so every yml-configured load put a dishonest agentOptions: {} on every start
request and the presence check in execute() could never be false through
config (only unit tests bypassing schemastery ever exercised that branch).
Harmless downstream today (the driver only spreads it), but the request
shape lied and the check was production-dead.
Same discipline as its toolFilter sibling: the omitted key now defaults to
undefined, the presence check is spelled !== undefined like its neighbors,
and a regression test (fails against the unfixed schema) pins that an
omitted agentOptions stays absent from the request. Swept every other
Config in the repo for the class: no further instances — omitted primitives
inside a materialized object stay ABSENT (verified empirically), so
subagent-mock's capabilities spread is safe, and the remaining object/array
fields all carry explicit defaults or the forced-undefined discipline
already.
Adversarial-review findings (own reviewer agent), each verified and pinned:
B1: agents.register() returned a wrapper lambda, so the factory composite's
yield could not identity-nest it — on OWNER unload the unregistration (and
agent/disposed) disposed as a concurrent sibling, firing mid-drain while
the final turn was still closing (pre-existing on master; this branch's
docs re-assert the order, so it must be true). register() now returns the
EXACT cordis effect disposer (the Scope.rawDispose move); the composite
nests it and owner unload runs stop/drain -> unregister -> detach -> scope
like every other path. Regression test pins turn-end before disposed
before detach on owner unload.
B2: the structured two-phase commit could promote a stale stage when a
later capture call REUSED the orphaned stage's call id with a body that
never staged (denied downstream, or invalid args throwing pre-stage). The
runtime's pre-execute listener now clears any stale stage unconditionally
when a new capture call enters the pipeline — only a call's own body can
stage for its commit; the call-id mismatch guard becomes a defensive
second layer. Repro test: blocked capture then same-id invalid call.
C1: an explicit empty toolFilter config now fails at plugin LOAD (the
check is self-contained) instead of killing every delegation at child
setup. C2: Scope.dispose/ScopeHost.dispose @returns state the single-shot
repeat-call semantics honestly.
ds-review-bot finding: forcing only the OUTER toolFilter key absent left
the nested arrays materializing — a deny-only config gained allow: [],
which means deny-EVERYTHING. The nested arrays now default to undefined
too; an explicit allow: [] (grant-only children) still survives. Pinned by
a capture-provider regression test.
Every subject-extractor row of the invariants carrier table is exercised
with a matching and a foreign-keyed carrier; the HMR re-apply seed path
(sessions of agents that predate the plugin are marked started) is pinned;
the scoped tool-provider disposal, plural restrict() validation, singular
scopeHost absentee, tool-subagent passthrough, stale-stage drop, and
disposing-parent spawn (INACTIVE_EFFECT, no orphan) each gain their test.
Two genuinely defensive branches carry justified v8-ignore markers.
SubagentStartRequest gains persona (capability-gated like toolFilter); the
in-process driver composes the child's scoped world in the factory's setup
window — persona as a scoped shadowing deployment:persona section,
toolFilter as a scoped tools.restrict() (loud unknown-name validation),
outputSchema as the scoped structured runtime. spawn/fork now advertise
every start-time capability; ACP stays all-false. A parent-scope teardown
effect links each child to its parent through the memoized handle, so a
disposed parent reaches its whole subtree even if the delegating tool's
finally never runs; subagent/start|end dispatch in the delegating parent's
scope.
structured.ts loses the placeholder schema, the final-assembly swap/strip,
the refcounted root runtime, and the WeakMap state: each child registers
its OWN capture tool (real schema), instruction section, and enforcement
listeners on child.ctx, riding the child's fiber. The commit listener is
call-keyed (a stale stage from a short-circuited post-execute chain is
dropped, never promoted on a later call), and one scoped prepend re-assert
listener preserves the final-assembly guarantee against a stripping global
listener.
tool-subagent gains persona/toolFilter/maxDepth passthrough config —
deny-listing the delegation tool (or maxDepth) is how a deployment bounds
recursion; the omitted-toolFilter schema key is forced absent (a
materialized {} would mean an empty allow-list, i.e. deny-everything).
Review round 2 (tianyicui inline comments):
- dsh-system-prompt itself registers the harness:identity (-100) and
deployment:persona (0) sections — they must survive a swapped loop
plugin, so they leave dsh-agent-loop; the persona text is the plugin's
own validated 'persona' config. The model/cwd variables STAY on the
loop: runtime facts of the agents it drives.
- AgentOptions.systemPrompt is deleted with all its forwarding plumbing:
the app configs' systemPrompt keys become 'persona' routed through
dsh-agent-core (schema = z.intersect of the owners'), the ACP bridge
and tool-subagent stop carrying persona configuration, and subagent
children now render the deployment persona like every other agent.
- Example personas drop transport/interface trivia (ACP, CLI) — facts
irrelevant to the model.
- Root CONTEXT.md removed (not idiomatic); its persona definition was
wrong under the new ownership anyway.
- Docs, READMEs, the prompt-variables RFC, and generated catalogs
updated; new loop test pins the assemble-waterfall escape valve
(an emptied assembly sends NO system field).
Codex round-2 findings:
- A committed test now proves the REAL plugin fiber (not a direct apply)
owns the provider-lifecycle listeners: disposing a mounted tool's fiber
unmounts the tool and leaves the provider intact, and a fiber disposed
while WAITING never zombie-mounts when its provider arrives later.
- TODO(subagent-dup-toolname) records the invalid-config blast radius of
two waiting fibers sharing a toolName (the duplicate throw propagates
through subagent/provider-added and rolls back the provider).
- CONTEXT.md drops its creation-history sentence; the RFC's acceptance
checklist becomes present-tense shipped invariants (docs/AGENTS.md
writing rules).
Codex round-1 findings, both confirmed:
- renderPrompt: variable lookup now uses Object.hasOwn (an unregistered
{{constructor}} previously resolved through Object.prototype and spliced
function source into the prompt), and a {{ that opens no complete group
while a }} still follows ({{{model}}}, {{a{b}}) now throws instead of
passing or partially interpolating. A lone {{ with no }} after it stays
verbatim; substituted values are never re-scanned.
- tool-subagent: the apply-time provider lookup assumed a load order the
cordis Loader does not guarantee (siblings start concurrently). The seam
now announces subagent/provider-added/-removed and the tool mirrors the
provider's lifecycle: registers when the provider is (or becomes)
available, unregisters when it goes away, re-derives wording on reload.
No load-order requirement remains.
- loop.spec containment test now proves live continuation: after the
contained render failure, a waterfall listener rescues {{cwd}} and the
same agent completes a real model turn.
RFC/READMEs updated to the shipped contract; cordis catalog regenerated.
One principle: every fact in the assembled prompt has exactly one owner.
- dsh-system-prompt: merge-extensible AssembleContext on assemble();
a variable(name, provider) registry; {{name}} interpolation in
renderPrompt, strict (unknown/valueless/malformed references throw);
duplicate section and variable names rejected; assembly carries
resolved section text + variables through the assemble waterfall.
- dsh-agent declares AssembleContext.agent; dsh-agent-loop registers
the agent:persona section (order 0 - identity renders before tool
guidance) and the model/cwd variables, and drops its string join:
renderPrompt(assembly) IS the full prompt.
- Tool guidance moves to its owners: descriptions carry per-tool
semantics; sections only cross-call habits (tool:bash exit-code
habit at order 105; read's not-shell nudge). todo/subagent need no
section - their descriptions already carry the contract.
- SubagentProvider.inheritsParentContext (spawn/acp false, fork true);
dsh-tool-subagent derives truthful per-provider wording and resolves
the provider at load (backend must be listed first).
- Example personas shrink to identity + behavior with {{model}} (and
{{cwd}} in the ACP tree); the welcome banner stops enumerating tools.
RFC: docs/rfc/implemented/architecture/2026-07-05-prompt-variables-and-tool-guidance-ownership.md
Reshape the agent's interception surface so every seam returns a small, typed
Decision union, and the set covers the hook points a CC/Codex bridge (and a
native plugin) needs. "Native hooks" are not a package — a native hook is just a
cordis plugin on these canonical events; the bridges (a later PR) only translate
an external protocol onto the same surface.
dsh-agent:
- NEW agent/session-start(agent, source) emit (once before turn 1; SessionStartSource
startup|resume|clear|compact) — a pure notification, seeds context via inject().
- NEW agent/prompt-submit waterfall → PromptDecision (allow, optionally rewriting the
prompt or attaching additionalContext, or block).
- RESHAPE agent/turn-continuation boolean → ContinuationDecision ({action:'stop'} |
{action:'continue', reason?}; a continue reason is recorded as next-step steering).
- New HookContext envelope (required source — inject() would mislabel a missing one).
dsh-tools: split the single tools/execute waterfall into tools/pre-execute
(PreToolDecision allow/deny/ask gate) and tools/post-execute (PostToolDecision
accept/block, optionally replacing content or attaching additionalContext). Core
dispatch sits between as plain code; the tool body keeps its inner try/catch so a
thrown tool still reaches post-execute as an isError. ToolExecutionResult gains
additionalContext (ferried to the loop's per-step buffer). Input rewrite is
deliberately NOT offered (a proposed RFC designs it consistently).
dsh-session: new `rejected` TurnEndReason — a turn whose whole prompt batch was
blocked by prompt-submit.
agent-loop firing points: session-start emitted at create (source threaded —
startup for create/fork, resume for resume()); prompt-submit per drained message
with the always-open-turn rule (a fully-blocked batch is a zero-step rejected
turn); the continuation reshape; post-tool additionalContext buffered and appended
after all tool/results (adjacency). ACP codec maps rejected→cancelled.
A worked native-plugin example (interception.spec.ts) proves all four seams compose
end-to-end through the real loop with NO hook/* events (those belong to the bridge
lib). All existing tools/execute + turn-continuation tests migrated. The
tool-subagent abort test now aborts after a microtask so it still exercises the
live onAbort bridge (execute() awaits pre-execute before the body runs).
RFCs: implemented/feature/2026-06-30-interception-seams.md (the reshape) +
proposed/feature/2026-06-30-pre-tool-input-rewrite.md (the deferred rewrite design).
addEventListener('abort') does not fire for a signal already aborted before the
listener is added, so a parent step cancelled before the subagent tool ran
would never reach the child — the tool leaned on each provider re-checking
request.signal itself, leaving the bridge's own claim incomplete for any
provider that relies on run.cancel(). Re-check exec.signal.aborted right after
registering and cancel explicitly. Regression test uses a spy provider that
only reacts to cancel() (never inspects the signal); proven to hang without the
fix (result never settles) and settle aborted with it.
Address four findings from the first Codex review round:
- Contain subagent/start|end listener throws (emitContainedStart/End): a
thrown lifecycle listener could escape SubagentService.start() before the
caller received the live run to dispose it (a leaked child), and a thrown
subagent/end listener could surface as an unhandled rejection on the detached
result-settle hook. Both emits now log-and-contain, mirroring the agent
registry's agent/created|disposed containment.
- Make the model-facing tool name configurable (Config.toolName, default
subagent). The docs say to load dsh-tool-subagent once per provider to expose
multiple transports, but the hardcoded name made the second load throw a
duplicate-tool-name error; a distinct toolName per load is now required and
documented.
- Reach the per-file 100% coverage gate: tests for the subagent/end error
branch, lifecycle-listener containment, every stopReasonError arm + the
merge-extensible default, the multi-provider toolName path, agentOptions
forwarding, and the direct-apply schema-bypass fallbacks.
- Document the seam vocabulary in docs/core-data-structures/subagent.md with
verbatim type-equiv blocks + manifest entries, and link it from core.md (a
brand-new core/seam type the doc-sync gate cannot detect on its own).
Introduce the `packages/subagent/` group and the abstract subagent seam — an
agent delegating to a child agent — as a named-provider registry (`ctx.subagents`),
unlike the single-implementation bash seam, so multiple transports (in-process,
ACP, future A2A) coexist. This first PR lands the interface, a scripted test
backend, and the model-facing tool, validated through the real cordis load path.
- dsh-subagent: SubagentService registry + SubagentProvider/SubagentRun
vocabulary + subagent/start|end events. Start-time capabilities (outputSchema,
depthLimit, toolFilter) are checked pre-start and rejected loud; runtime
capabilities (sendMessage, resume) are optional methods on SubagentRun.
- dsh-subagent-mock (support): scripted provider for keyless, deterministic
tests through the real Loader/export path.
- dsh-tool-subagent: the model-facing `subagent` tool, config-bound to one
provider; synchronous collect with try/finally dispose, signal->cancel
bridging, and non-completed-stop-reason -> isError mapping.
- Proposed RFC documenting the seam, the fork-vs-spawn-as-separate-backends
decision, own-session isolation, synchronous-collect scope, and the deferral
of background/poll/spill to a future unification with bash.
- Wire the new group into tsconfigs, build refs, package hierarchy docs, the
module graph, and the cordis catalog.
RFC: docs/rfc/proposed/feature/2026-06-21-subagent-capability-seam.md
Tianyi Cui
Yichen Jiang
Dudu-0223