Reconcile the session-surface feature with master's package reorg and
simplifications:
- Adopt master's folded usage (assistant/message.usage; standalone `usage`
event dropped) and re-attach surface metadata (surfaceOp/sourceEventSeqs).
- Add surface opts to master's new max-tokens assistant/message append.
- Port surface columns onto the coordinator-refactored SQLite backend at its
new path; drop the dead v1->v2 migration (bump-and-reject, no migration per
pre-release policy).
- Move the session-surface RFC into implemented/architecture/ and refresh its
stale body (no migration, SESSION_FORMAT_VERSION=0, renamed package paths).
- Update the core-data-structures catalog SessionEvent blocks for the two new
surface fields; regenerate the cordis catalog.
- Re-harvest ACP snapshot fixtures (keyless replay) to carry surface metadata.
The createdAt+recordedId child sort comment over-claimed "tie-safe". Codex
flagged that a same-millisecond sibling tie would be broken by random session
id, which does not recover first-call order. In the current synchronous cut that
tie is unreachable — the subagent tool awaits one child's result and disposes it
before the parent starts the next, so siblings' createdAt values are strictly
ordered and match first-call order. Restate the comment to that real invariant
(at both the replay sort and the harvest sort), note that the id tiebreak only
makes a degenerate collision deterministic, and flag the concurrent-subagent cut
that would need a real first-call ordinal with XXX(concurrent-subagents). The RFC
records the same limitation. Comment/doc only — no behavior change.
The snapshot tier was built single-session: dsh-llm-replay served calls from
one global positional cursor, and the harness harvested one session log. A
subagent runs as a second agent with its own session, so a parent→child
scenario could neither replay deterministically nor harvest the child's log.
This resolves the TODO(subagent-snapshots) deferral from the subagent RFC.
- Stamp the calling session id onto the model request: GenerateOptions.sessionId
(typed Branded<'SessionId'> to avoid the dsh-llm↔dsh-session cycle), set by the
agent loop from agent.session.id. Adapters ignore it; an llm/stream listener
routes by it.
- Key replay per session: dsh-llm-replay loads the parent log plus one per child
(childFiles / $DSH_SNAPSHOT_CHILD_FILES), derives a script per recorded session,
and binds each live (freshly-random) session to a recorded script by first-call
order — parent first (earliest createdAt, first to stream). Keys by WHO calls,
so it survives a future concurrent/backgrounded subagent; a global cursor would
not. An unrecorded extra session fails loud.
- Harvest every log: the harness collects all .jsonl across cwd buckets, ordered
primary-first (top-level, then children by createdAt), and RunResult exposes the
plural sessionLogs. The spec writes each back on record (session.jsonl +
session.<n>.jsonl) and diffs each against its fixture on replay.
- Wire the subagent seam + spawn + fork + tool into the acp-agent example (both
cordis configs) and add two nested scenarios recorded against the real API:
subagent-spawn (parent + 1 child) and subagent-multi (parent + 2 children, 3
sessions). Both replay keyless in the default gate.
A new RFC documents the design (docs/rfc/implemented/testing/). Single-session
replay is unchanged (a call with no sessionId is one anonymous primary session).
TODO follow-up: a dedicated branded-ids package could own the SessionId brand and
dissolve the cross-package cycle note; out of scope for this testing PR.
A request signal aborted BEFORE the run starts never fires an `abort` event
(`addEventListener` only fires on the transition), so the backend-level bridge
missed it and ran the child to `completed`. The driver now checks
`request.signal?.aborted` at the top of the result path and settles `aborted`
without running the child. Regression test proven red on the pre-fix code.
Also refresh two stale RFC prose blocks the round-1 fix left behind: the
subagent RFC's Problem statement (cited the removed `TODO(sub-agents)` markers
and claimed nothing existed yet) and the unify-id RFC's fork/spawn risk bullet
(described the seam as "explicitly deferred" via `AgentLoop.create`'s old TODO),
now pointing at the realized seam.
Two merge-blocking bugs in the shared in-process run driver, both rooted in
`readResult` scanning the whole child session and deriving the stop reason only
from `turn/end`:
- A pre-turn `cancel()` cleared the queued prompt before any `turn/end` was
logged, so the run settled `error` instead of `aborted`, violating the
`SubagentRun.cancel()` contract. The driver now tracks that a cancel was
requested and maps the no-turn case to `aborted`.
- A fork child whose own turn produced no `assistant/message` returned the
SEEDED parent's last message as a `completed` success. `readResult` now scopes
to the child's OWN events (after the seed prefix), so a message-less child
yields empty output.
Both fixes carry a regression test proven to go red on the pre-fix driver.
Also: correct the `SubagentRun.id` / event-payload docs (it is the child AGENT
id, not a session id — the backend mints distinct tokens); refresh the stale
`coding-agent` welcome string (subagent is now a tool); and replace the stale
`TODO(sub-agents)` "deferred" prose in the Agent interface, core.md, and
architecture.md with an accurate pointer to the realized seam.
The second PR of the subagent seam: the two in-process backends that run a
child agent on the same cordis context, reusing the agent factory's quiescent
AgentHandle teardown. Both register on ctx.subagents (PR1's named-provider
registry) and share one run driver.
- dsh-subagent-spawn: a FRESH child via ctx.agents.create — own session, the
parent's model by default (overridable), zero inherited conversation. Also
exports the shared in-process run driver (startInProcessRun): mint ids, stamp
cwd/parentSession-lineage/depth, drive the one-shot (send → whenIdle), read
the last assistant/message + turn/end reason, dispose to quiescence.
- dsh-subagent-fork: a child SEEDED with the parent's balanced completed-turn
prefix (the log up to and including its last turn/end), so the child inherits
context. The in-flight unbalanced turn is excluded — a raw seed would fail the
invariants replay. Proven: a regression test goes red if the boundary seeds
the open turn.
- Seam extension: CreateAgentOptions.seed, threaded through AgentLoop.createAgent
→ ctx.sessions.prepare({ seed }) (the primitive resume already used). This is
the fork-lineage path the TODO(sub-agents) markers anticipated.
- Depth: a merge-extensible AgentOptions.subagentDepth (0 top-level, parent+1 for
a child); the depthLimit capability refuses a spawn past request.maxDepth.
Tests: real-loop unit tests for both backends (mock MODEL only, real loop +
invariants), a multi-subagent test (one parent drives a fork AND a spawn child
then keeps working), and a with-key e2e (a real parent delegates via the
`subagent` tool to a real child that writes a file on disk — world-verified).
100% per-file coverage. The coding-agent demo wires the spawn backend + tool.
Snapshot coverage of nested agents is deferred to a stacked follow-up
(TODO(subagent-snapshots)): dsh-llm-replay is a single global positional cursor
that cannot route calls to a parent vs. a child on one context. Recorded in the
RFC's deferrals and a new AGENTS.md rule: designing a subsystem must design its
test infrastructure END TO END up front, verifying the snapshot/e2e harness can
express the new shape — a gap this plan hit.
A single try/catch around ctx.emit prevented a thrown subagent/start or
subagent/end listener from propagating, but cordis emit dispatches listeners in
a `.map(cb => cb())` that HALTS on the first throw — so a bad subscriber still
starved the listeners registered after it, violating the AGENTS.md
callback-boundary rule ("one bad subscriber must not starve the listeners after
it"). Resolve the listener callbacks via ctx.events.dispatch and contain each
call individually, the same per-listener guarantee BashExecutor.notifyTaskDone
gives its own listener set.
The two containment tests now register TWO listeners where the first throws and
assert the second still observes the event (start) and the settle (end) — a
regression that fails on the per-emit code (verified: reverted, watched both go
red, restored).
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
assertEntriesLoaded() flagged ANY fiber-less entry as a failed import, but a
`disabled: true` entry settles without a fiber by design (Entry.refresh() skips
init() when disabled) — a valid "plugin off" config, not a broken import. Both
app bins now filter `fiber === undefined && !entry.disabled`. The stdio built-bin
smoke gains a disabled-(unresolvable)-entry config that must still boot.
The verify-package-paths lib-skip was unconditional and ran before the
moved-package check, so a stale group-less `packages/acp-agent/lib/bin.js` (the
exact drift this gate catches) was silently ignored just for containing `lib`.
Scope the skip: only exempt `lib` when it is the segment after an EXISTING
`packages/<group>/<pkg>` root, so a real-but-unbuilt `lib/bin.js` is still exempt
while a stale package path flags.
The required-fixture-guard description still said session.jsonl was needed only
for model scenarios, but the harness passes <dir>/session.jsonl to llm-replay
unconditionally, so loadReplayScript() fails for a no-model scenario without it.
The code already requires it for all scenarios; align the RFC prose.
The retained-whenIdle paragraph claimed "live consumers (the ACP bridge's settle
points)", but `packages/ui/acp/src` has no whenIdle() call — the bridge owns its
agents and tears them down via AgentHandle.dispose(). whenIdle()'s live consumers
are ACP and agent TESTS awaiting settlement through the public seam. State that.
verify-package-paths flagged the new built-bin smokes' `lib/bin.js` citations
as stale-source drift, failing CI: doc-sync runs BEFORE build, so the build
output is absent at lint time. The gate targets moved SOURCE paths, so skip any
reference whose target goes through a `lib/` segment — mirroring how the file
scan already excludes `lib/`.
The acp built-bin smoke resolved `zod`/`@agentclientprotocol/sdk` via
`import.meta.resolve` from the test file's own context, but `acp-agent` does not
declare them — `dsh-acp` does. Under pnpm's strict layout they are not exposed
where the test resolves, so the new CI built-bin step failed with
"Cannot find package 'zod'". Resolve each from the `ui/acp` package URL (the one
that declares it) instead.
Three hard-won lessons from running a 7-PR simplification stack through two
waves of review feedback:
- New "## Orchestrating review feedback across a stacked PR chain" section: one
worktree per branch; a fix belongs on the PR that introduced the issue then
flows DOWN; review-fixes are separate commits never amends; delegated work is
trust-but-verify (prove a regression guard FAILS on unfixed code); reply
in-thread on the merits.
- "## Conventions" gains a "Never rewrite a pushed branch" rule next to the
merge-commit rule: update a child by merging the parent down, never
rebase/amend/force-push a pushed branch; a fix lands on its originating PR.
- Extended the "Line coverage is not behavior coverage" defensive-patterns
bullet with two corollaries this stack re-taught: (1) a real-load-path test
only GUARDS the export shape if a broken shape actually FAILS it — an
inject-less composition plugin boots fine on a stray `export default`, so it
needs an explicit no-default + unwrapExports assertion; (2) "real entry path"
means the PUBLISHED artifact (built lib/bin.js under plain node), not the dev
runtime (tsx), which masks boot settle-races, module-resolution differences,
and a load failure that loader.await()'s Promise.allSettled swallows.
The per-kind fixture guard claimed no-model scenarios need no session.jsonl, but
the replay path requires one for ALL scenarios: runScenario() passes
`fixtureFile: <dir>/session.jsonl` unconditionally and llm-replay's
loadReplayScript() throws "fixture not found" when it is absent and no override
replaces it. A no-model scenario ships a header-only session.jsonl that derives
to an empty script. The guard + its comment now match that reality, so a future
no-model scenario following them won't fail at subprocess startup.
