- turn boundary now returns false for an empty admitted batch (claimed
input removed before the wake) instead of opening a turn and spending a
model call on nothing; the step boundary already had the symmetric guard.
- a max-token step stays sticky when steering or injected work continues
the turn: a later completed step no longer downgrades the outcome,
matching the TurnEndReasonMap contract.
- session/queue wire schema accepts the context placement (previously the
zod union rejected injected-context snapshots wholesale and the client
silently dropped the whole frame); schema tests cover all placements.
- headless runs settle at whole-agent idle instead of the first turn/end,
honoring the one-shot idle-to-idle contract.
- flush JSDoc names the real callers (checkpoint policy, goal-session,
teardown, self-flushing consumers); apiproxy zh README loses its stale
duplicate history section; ACP note/README record the delivered error
rejection and turnless-cancelled behaviors.
Appending todo/write, request/header, and request/context inside an open
turn now exercises the switch cases and the enclosed-branch of the
invariant, closing the last branch gap.
- invariant.spec: a turn/end whose step does not match the last opened
step fails with 'expected last step'.
- session.spec: a seed message whose content is not an array fails with
'message has invalid content'.
- gen-cordis-catalog / gen-persistence-catalog / gen-doc-graphs after the
error-reason and steering/message changes (line anchors and event
producer/consumer graph drifted).
- verify-translation-pairing --write --all: every bilingual pair touched
by the PR (agent notes, session/core docs, READMEs) plus the merge.
- queue-actions preserved golden re-recorded against the rebuilt client
(branch disabled on non-completed-turn tails is back in effect).
TurnEndReasonMap.error now carries a single `error: LlmFailure` field:
an LlmError keeps its structured facts, any other error flattens to
errorChain text under the UNKNOWN code. Consumers read message/code
directly instead of defending against an unknown union — this also fixes
errorChain() rendering structured failures as '[object Object]' in the
TUI and ACP error paths. Document the turn-stopping contract: a
concludesTurn result never short-circuits already-submitted next-step
work (same-step additionalContexts or racing steering still runs), data
decides.
A prompt now resolves only when the agent reaches quiescence: the
correlated turn/end arms an endReason instead of settling immediately,
so steering or injected work that runs further turns before idle no
longer splits one prompt across the automation wire. Token-limit turn
endings settle as end_turn per the README contract (they are not
prompt-level stop reasons); model errors still reject immediately, and
explicit ACP cancellation stays 'cancelled'.
next-step items now split by origin: user-origin messages keep the
'steering' placement, while agent.inject context (approval notices, task
completion notices, attached snapshots) carries a new 'context' placement
that no surface renders until it is claimed as a durable user/message
context card. Widen the placement unions on the wire and in the client
runtime, add a projection test case, and document the split in the
apiproxy/ui-conversation READMEs and the web-steer chrome agent note.
- agent/error: drop the stale "(plus the logger)" claim; the machine only
emits the event.
- agent/inbox/claimed: document that a rejected turn boundary ends the
claimed message without discard or re-emission, and that a later batch
may reuse the turn number.
- steer(): document that a rejected step leaves steering parked until the
next wake.
- Fix stray double-space in the inbox inserted/discarded callbacks.
- Sync core.md/core.zh.md mirrors and regenerate the cordis catalog.
Steer, inject, and followup now land as durable user/message events on the
session surface; the steering/message event type and its ConversationNode
kind are removed from the client projection. Update tests, docs, generated
catalogs, and agent notes to match, and align the steering e2e fixture and
prompt inventory assertions with the durable user/message landing.