docs(agent): mark claim as the loop-internal step-boundary operation

packages/core/agent/README.i18n.yaml

@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/core/agent/README.md
-README.md: a3781f1ab466e6825a80c529c261b3a4ea262092
-README.zh.md: 77a082f3ff83caf4f3934156e4546a9a227d42cc
+README.md: c3d6e6c24480894b6059417c1ab89db7aa0d7fa2
+README.zh.md: 16ee8f5e6c483555839b0c3ab174e2e2356b1359

packages/core/agent/README.md

@@ -62,7 +62,7 @@ Turn and step boundaries and the model token stream are durable `session/event`
 
 The handle every plugin programs against:
 
-- `agent.inbox` — the agent-owned projection of durable `agent/inbox/spliced` events. `nextTurn` and `nextStep` expose pending `UserMessage` values. `append`, `prepend`, `replace`, `remove`, `clear`, and `splice` mutate them; `replace(messageId, newMessage)` and `remove(messageId)` locate the pending message across both lists. Replacement may change identity and publishes the old message as discarded followed by the new message as inserted. Ordinary removals and `clear()` are durable cancellations and emit `agent/inbox/discarded`. `claim(target)` atomically removes the next proposed batch with pure deletion splices; the loop then emits `agent/inbox/claimed`. `MessageId` is the only occurrence identity and must remain unique while pending.
+- `agent.inbox` — the agent-owned projection of durable `agent/inbox/spliced` events. `nextTurn` and `nextStep` expose pending `UserMessage` values. `append`, `prepend`, `replace`, `remove`, `clear`, `splice`, and `claim` mutate them; `replace(messageId, newMessage)` and `remove(messageId)` locate the pending message across both lists. Replacement may change identity and publishes the old message as discarded followed by the new message as inserted. Ordinary removals and `clear()` are durable cancellations and emit `agent/inbox/discarded`. `claim(target)` removes the next proposed batch with pure deletion splices; the loop then emits `agent/inbox/claimed`. `MessageId` is the only occurrence identity and must remain unique while pending.
 - `agent.followup(message)` — queue an ordinary `next-turn` message and wake the driver. It returns no completion handle; the message id identifies inbox insertion, claim, and discard facts, not a later output or `turn/end`.
 - `agent.steer(message)` — queue waking `next-step` input. An idle agent starts a turn synchronously; a running driver consumes later steering at its next step boundary.
 - `agent.inject(message)` — queue non-waking `next-step` context. A running driver claims it at the nearest later pre-step boundary; an idle driver leaves it pending until `followup()` or `steer()` wakes the driver. It may miss a request whose pre-step already claimed its batch.

packages/core/agent/README.zh.md

@@ -62,7 +62,7 @@ inbox 的实时通知刻意采用逐消息的最小载荷：`agent/inbox/inserte
 
 每个插件面向的 handle：
 
-- `agent.inbox`：agent 所拥有的持久 `agent/inbox/spliced` 事件投影。`nextTurn` 与 `nextStep` 暴露待处理的 `UserMessage` 值。`append`、`prepend`、`replace`、`remove`、`clear` 与 `splice` 用于变更队列；`replace(messageId, newMessage)` 与 `remove(messageId)` 通过 `MessageId` 跨两份列表定位待处理消息。替换可以改变标识，并先将旧消息作为 discarded 发布，再将新消息作为 inserted 发布。普通删除和 `clear()` 都是持久取消，并发出 `agent/inbox/discarded`。`claim(target)` 通过纯删除 splice 原子移除下一个候选批次，随后由循环发出 `agent/inbox/claimed`。`MessageId` 是唯一的入队项标识，在消息待处理期间必须保持唯一。
+- `agent.inbox`：agent 所拥有的持久 `agent/inbox/spliced` 事件投影。`nextTurn` 与 `nextStep` 暴露待处理的 `UserMessage` 值。`append`、`prepend`、`replace`、`remove`、`clear`、`splice` 与 `claim` 用于变更队列；`replace(messageId, newMessage)` 与 `remove(messageId)` 通过 `MessageId` 跨两份列表定位待处理消息。替换可以改变标识，并先将旧消息作为 discarded 发布，再将新消息作为 inserted 发布。普通删除和 `clear()` 都是持久取消，并发出 `agent/inbox/discarded`。`claim(target)` 通过纯删除 splice 移除下一个候选批次，随后由循环发出 `agent/inbox/claimed`。`MessageId` 是唯一的入队项标识，在消息待处理期间必须保持唯一。
 - `agent.followup(message)`：将一条普通 `next-turn` 消息排队并唤醒驱动器。它不返回完成 handle；消息 id 标识 inbox 的插入、领取与丢弃事实，而不标识之后的输出或 `turn/end`。
 - `agent.steer(message)`：将会唤醒的 `next-step` steering（中途引导）输入排队。agent 空闲时会同步启动一个轮次；驱动器运行期间收到的后续 steering 会在下一个步骤边界被消费。
 - `agent.inject(message)`：将不会唤醒的 `next-step` 上下文排队。运行中的驱动器会在最近的后续 pre-step 边界领取它；idle 驱动器则会让它保持待处理，直至 `followup()` 或 `steer()` 唤醒驱动器。若某次请求的 pre-step 已经领取完批次，它可能赶不上该请求。

packages/core/agent/src/inbox.ts

@@ -65,6 +65,7 @@ export class Inbox {
    * splices are pure deletions; the caller publishes claimed notifications.
    * @param target - whether this boundary also consumes one queued turn.
    * @returns next-step input followed by the queued turn, when requested.
+   * @internal - the agent loop's step-boundary operation, not a plugin seam.
    */
   claim(target: InboxTarget): UserMessage[] {
     const claimed = this.mutate('next-step', 0, this.nextStep.length, [], false)