review: one fold implementation, uniform end-edge absence, partial text in tool errors, snapshot scenario

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.
2026-08-10 16:09:04 +08:00
parent 2418f44781
commit ea47c32805
24 changed files with 221 additions and 96 deletions
--- a/packages/subagent/subagent-acp/src/run.ts
+++ b/packages/subagent/subagent-acp/src/run.ts
@@ -24,6 +24,7 @@ import {
 } from '@agentclientprotocol/sdk'
 import type { ContentBlock } from '@deepseek-ai/dsh-llm'
 import { SessionId } from '@deepseek-ai/dsh-session'
+import { AssistantOutputFold } from '@deepseek-ai/dsh-subagent'
 import type { SubagentResult, SubagentRun, SubagentStartRequest, SubagentStopReason } from '@deepseek-ai/dsh-subagent'
 import type { SubprocessHandle, SubprocessSpawnSpec } from '@deepseek-ai/dsh-subprocess'

@@ -232,8 +233,10 @@ export async function startAcpRun(request: SubagentStartRequest, spec: AcpRunSpe
  let processDisposal: Promise<void> | undefined
  const disposeProcess = (): Promise<void> => (processDisposal ??= disposeAcpChild(child, spec.disposeEofGraceMs))

-  // Accumulate the child's streamed assistant text — the SubagentResult output.
-  const output: string[] = []
+  // The child's streamed assistant text, accumulated under the seam's
+  // canonical selection rule (`AssistantOutputFold`); ACP surfaces no complete
+  // assistant messages, so only the streamed-fallback half applies.
+  const fold = new AssistantOutputFold()
  // Shared mutable state keeps cancellation visible across async closures.
  const flags = { cancelled: false }

@@ -241,7 +244,7 @@ export async function startAcpRun(request: SubagentStartRequest, spec: AcpRunSpe
    sessionUpdate(params: SessionNotification): Promise<void> {
      const update = params.update
      if (update.sessionUpdate === 'agent_message_chunk') {
-        output.push(acpContentText(update.content))
+        fold.pushText(acpContentText(update.content))
      }
      // Other updates (thoughts, tool calls, plans) are consumed but not
      // surfaced — the subagent returns only its final answer.
@@ -284,13 +287,8 @@ export async function startAcpRun(request: SubagentStartRequest, spec: AcpRunSpe
  const onAbort = (): void => { requestCancel() }
  request.signal.addEventListener('abort', onAbort, { once: true })

-  // The accumulated child text as harness ContentBlocks (empty array when the
-  // child streamed nothing). Read at every return so a partial answer survives
-  // a later cancel/error.
-  const collectOutput = (): ContentBlock[] => {
-    const text = output.join('')
-    return text.length > 0 ? [{ type: 'text', text }] : []
-  }
+  // Read at every return so a partial answer survives a later cancel/error.
+  const collectOutput = (): ContentBlock[] => fold.collect() ?? []

  // Establish the remote session before publishing a handle. Any failure owns
  // the still-private process and therefore reaps it before rejecting.
--- a/packages/subagent/subagent-dsh-sdk/src/run.ts
+++ b/packages/subagent/subagent-dsh-sdk/src/run.ts
@@ -16,7 +16,7 @@ import { DeepSeekHarness, type HarnessNotification } from '@deepseek-ai/dsh-sdk-
 import type { ContentBlock } from '@deepseek-ai/dsh-llm'
 import { SessionId, type SessionEvent, type TurnEndReason } from '@deepseek-ai/dsh-session'
 import type { SubagentResult, SubagentRun, SubagentStartRequest, SubagentStopReason } from '@deepseek-ai/dsh-subagent'
-import { assistantMessageOutput, settleRunResult, subprocessRunHandle } from '@deepseek-ai/dsh-subagent'
+import { AssistantOutputFold, settleRunResult, subprocessRunHandle } from '@deepseek-ai/dsh-subagent'
 import { scrubbedParentEnv } from '@deepseek-ai/dsh-subprocess'

 /** Resolved spawn spec for an SDK runtime child process (no defaults — see Config). */
@@ -163,28 +163,14 @@ export async function startSdkRun(request: SubagentStartRequest, spec: SdkRunSpe
  }

  const childSessionId = `session-${randomUUID().replaceAll('-', '')}`
-  // The child's final answer, folded incrementally under the seam's canonical
-  // rule (`finalAssistantOutput`): the last NON-EMPTY complete assistant
-  // message when one exists, else the text streamed so far (a partial answer
-  // surviving cancel). An empty-content message hosts only usage (a max-tokens
-  // step that assembled no text blocks), so it never erases streamed text.
-  let lastMessage: ContentBlock[] | undefined
-  const partial: string[] = []
+  // The child's final answer under the seam's canonical selection rule
+  // (`AssistantOutputFold`); a partial answer survives cancel and error paths.
+  const fold = new AssistantOutputFold()
  const observe = (notification: HarnessNotification): void => {
    if (notification.method !== 'session.event' || notification.params.sessionId !== childSessionId) return
-    const event = notification.params.event as SessionEvent
-    const content = assistantMessageOutput(event)
-    if (content !== undefined) {
-      lastMessage = content
-    } else if (event.type === 'assistant/chunk' && event.data.chunk.type === 'text-delta') {
-      partial.push(event.data.chunk.text)
-    }
-  }
-  const collectOutput = (): ContentBlock[] => {
-    if (lastMessage !== undefined) return lastMessage
-    const text = partial.join('')
-    return text.length > 0 ? [{ type: 'text', text }] : []
+    fold.push(notification.params.event as SessionEvent)
  }
+  const collectOutput = (): ContentBlock[] => fold.collect() ?? []

  // Race the child turn against local cancellation; the shared settlement
  // flattens failures under the seam's never-reject contract.
--- a/packages/subagent/subagent-inprocess/src/index.ts
+++ b/packages/subagent/subagent-inprocess/src/index.ts
@@ -220,9 +220,7 @@ function readResult(
 ): SubagentResult {
  const own = child.session.events.slice(boundary)
  const lastEnd = findLastMessageTurnEnd(own)
-  // Canonical selection (`finalAssistantOutput`): the last non-empty assistant
-  // message, else the text streamed before cancel/error/truncation cut the
-  // turn short — an empty usage-only message never erases real output.
+  // The seam's canonical selection rule; a partial answer survives cancel and truncation.
  const output: ContentBlock[] = finalAssistantOutput(own) ?? []
  const recorded = toStopReason(lastEnd?.data.reason)
  // Disposal can tear the owner down before the loop records its ordinary
--- a/packages/subagent/subagent/README.i18n.yaml
+++ b/packages/subagent/subagent/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/subagent/subagent/README.md
-README.md: d2d5356fd82a47ecf5cd6b633e5338dde7047901
-README.zh.md: 2fdc3ae6e8376ef7c7aaf11c8e85dd909ef92d2c
+README.md: 843fd0af4a86ea3a10d4a4ee101aa05478a2300e
+README.zh.md: 497f2a928c8eaeff8ef0486f5344c3d257be7391
--- a/packages/subagent/subagent/README.md
+++ b/packages/subagent/subagent/README.md
@@ -56,7 +56,7 @@ The seam owns the depth vocabulary shared by Service providers and Consumers: th

 `provider.start(request): Promise<SubagentRun>` is the ownership-transfer boundary; the delegation tool also uses it inside its one-shot Task-backed background path. Before fulfillment, the provider owns setup and must cancel, roll back, and quiesce unpublished resources on every failure. After fulfillment, the caller owns the run and must call `dispose()` on every path; remaining prompt and turn work belongs to `SubagentRun.result`.

-`SubagentRun.result` resolves to `{ output, structured?, stopReason }`. Child-level failures resolve with a non-`completed` reason; only an infrastructure fault that the seam cannot represent may reject. `dispose()` is idempotent, cancels remaining work, and waits for both result settlement and child-resource quiescence. A result rejection remains on `result`; `dispose()` rejects only for an independent resource-release failure. `output` and the `subagent/end` edge's `lastAssistantMessage` share one selection rule, implemented by the exported `finalAssistantOutput` helper: the child's last non-empty assistant message, else the text it streamed before the turn was cut short ([`SubagentResult.output`](../../../docs/subsystems/subagent.md#the-terminal-result-subagentresult) owns the contract).
+`SubagentRun.result` resolves to `{ output, structured?, stopReason }`. Child-level failures resolve with a non-`completed` reason; only an infrastructure fault that the seam cannot represent may reject. `dispose()` is idempotent, cancels remaining work, and waits for both result settlement and child-resource quiescence. A result rejection remains on `result`; `dispose()` rejects only for an independent resource-release failure. `output` and the `subagent/end` edge's `lastAssistantMessage` share one selection rule, implemented once by the exported `AssistantOutputFold`/`finalAssistantOutput` helpers: the child's last non-empty assistant message, else the text it streamed before the turn was cut short ([`SubagentResult.output`](../../../docs/subsystems/subagent.md#the-terminal-result-subagentresult) owns the contract).

 A local run publishes an ordinary child agent/session before `start()` fulfills, returns that shared session id as `SubagentRun.id`, exposes the exact child as `SubagentRun.localAgent`, records `request.parent.session.id` in the child's `parentSession` header, and appends the resolved descriptor inside its initial turn. Remote providers instead mint a parent-scoped lifecycle id and return `localAgent: undefined`; without a local child session, their one-shot runs are not part of trace-backed enumeration.

--- a/packages/subagent/subagent/README.zh.md
+++ b/packages/subagent/subagent/README.zh.md
@@ -56,7 +56,7 @@ subagent seam 允许一个 agent（智能体）通过具名提供方把工作委

 `provider.start(request): Promise<SubagentRun>` 是所有权转移边界；委派工具也会在其由 Task 支撑的一次性后台路径中使用它。兑现前，提供方拥有设置过程，并且每次失败时都必须取消、回滚并使未发布资源完全停稳。兑现后，调用方拥有该运行，并且必须在每条路径上调用 `dispose()`；剩余提示词和轮次工作属于 `SubagentRun.result`。

-`SubagentRun.result` 兑现为 `{ output, structured?, stopReason }`。子 agent 级失败会以非 `completed` 原因兑现；只有 seam 无法表示的基础设施故障才可以拒绝。`dispose()` 是幂等的，会取消剩余工作，并等待结果结算以及子 agent 资源完全停稳。`result` 的 rejection 仍归 `result` 通道；只有独立的资源释放失败会使 `dispose()` 拒绝。`output` 与 `subagent/end` 边沿的 `lastAssistantMessage` 共用同一条选取规则，由导出的 `finalAssistantOutput` 辅助函数实现：取子 agent 最后一条非空 assistant 消息，否则取轮次被截断前已流式的文本（契约归 [`SubagentResult.output`](../../../docs/subsystems/subagent.md#the-terminal-result-subagentresult) 所有）。
+`SubagentRun.result` 兑现为 `{ output, structured?, stopReason }`。子 agent 级失败会以非 `completed` 原因兑现；只有 seam 无法表示的基础设施故障才可以拒绝。`dispose()` 是幂等的，会取消剩余工作，并等待结果结算以及子 agent 资源完全停稳。`result` 的 rejection 仍归 `result` 通道；只有独立的资源释放失败会使 `dispose()` 拒绝。`output` 与 `subagent/end` 边沿的 `lastAssistantMessage` 共用同一条选取规则，由导出的 `AssistantOutputFold`／`finalAssistantOutput` 辅助函数唯一实现：取子 agent 最后一条非空 assistant 消息，否则取轮次被截断前已流式的文本（契约归 [`SubagentResult.output`](../../../docs/subsystems/subagent.md#the-terminal-result-subagentresult) 所有）。

 本地运行会在 `start()` 兑现前发布普通的子 agent／会话，把该共享会话 id 作为 `SubagentRun.id` 返回，以 `SubagentRun.localAgent` 公开准确的子 agent，把 `request.parent.session.id` 记录到子 agent 的 `parentSession` header，并在其初始轮次内追加已解析的描述符。远程提供方则生成 parent 作用域的生命周期 id，并返回 `localAgent: undefined`；由于没有本地 child 会话，其一次性运行不会进入基于追踪的枚举结果。

--- a/packages/subagent/subagent/src/assistant-output.ts
+++ b/packages/subagent/subagent/src/assistant-output.ts
@@ -1,12 +1,12 @@
 /**
- * Canonical selection of a child's final assistant output from its session
- * events. Every surface that reports "the child's answer" — backend run
- * results and `subagent/end.lastAssistantMessage` — applies this one rule so
- * observers agree: the last NON-EMPTY assistant message wins; an empty-content
- * message hosts only usage (the loop appends one when a max-tokens step
- * assembled no executable blocks) and never erases real output; without any
- * non-empty message, the text streamed so far is the answer (a partial
- * surviving cancel, error, and truncation paths).
+ * Canonical selection of a child's final assistant output. Every surface that
+ * reports "the child's answer" — backend run results and
+ * `subagent/end.lastAssistantMessage` — applies this one rule so observers
+ * agree: the last NON-EMPTY assistant message wins; an empty-content message
+ * hosts only usage (the loop appends one when a max-tokens step assembled no
+ * executable blocks) and never erases real output; without any non-empty
+ * message, the text streamed so far is the answer (a partial surviving
+ * cancel, error, and truncation paths).
 *
 * @module @deepseek-ai/dsh-subagent/assistant-output
 */
@@ -15,37 +15,57 @@ import type { ContentBlock } from '@deepseek-ai/dsh-llm'
 import type { SessionEvent } from '@deepseek-ai/dsh-session'

 /**
- * The content one event contributes as a candidate final answer: an
- * `assistant/message` with non-empty content. An empty-content message hosts
- * only usage and contributes none.
- * @param event - any session event.
- * @returns the message content, or `undefined` when this event is not a
- *   non-empty assistant message.
+ * Incremental fold of the selection rule, for backends that observe a child's
+ * output as it streams: session-event backends {@link push} each event, and
+ * transports without session events (ACP content chunks) {@link pushText} raw
+ * text into the same streamed fallback.
 */
-export function assistantMessageOutput(event: SessionEvent): ContentBlock[] | undefined {
-  if (event.type !== 'assistant/message') return undefined
-  const content = event.data.message.content
-  return content.length > 0 ? content : undefined
+export class AssistantOutputFold {
+  private message: ContentBlock[] | undefined
+  private partial: string[] = []
+
+  /**
+   * Fold one session event: a non-empty assistant message becomes the
+   * candidate final answer, and a `text-delta` chunk extends the streamed
+   * fallback; every other event contributes nothing.
+   * @param event - the next observed session event.
+   */
+  push(event: SessionEvent): void {
+    if (event.type === 'assistant/message') {
+      const content = event.data.message.content
+      if (content.length > 0) this.message = content
+    } else if (event.type === 'assistant/chunk' && event.data.chunk.type === 'text-delta') {
+      this.partial.push(event.data.chunk.text)
+    }
+  }
+
+  /**
+   * Extend the streamed fallback with text observed outside session events.
+   * @param text - the next streamed text piece (an empty piece is a no-op).
+   */
+  pushText(text: string): void {
+    this.partial.push(text)
+  }
+
+  /**
+   * Select the final output folded so far.
+   * @returns the last non-empty assistant message, else the accumulated
+   *   streamed text, or `undefined` when the child produced neither.
+   */
+  collect(): ContentBlock[] | undefined {
+    if (this.message !== undefined) return this.message
+    const text = this.partial.join('')
+    return text.length > 0 ? [{ type: 'text', text }] : undefined
+  }
 }

 /**
- * Select the final assistant output from one child-owned event suffix: the
- * last non-empty assistant message, else the accumulated `text-delta` stream.
+ * Apply the selection rule to one complete child-owned event suffix.
 * @param events - the child-owned events (after any seed or epoch boundary).
 * @returns the selected output, or `undefined` when the child produced none.
 */
 export function finalAssistantOutput(events: readonly SessionEvent[]): ContentBlock[] | undefined {
-  let message: ContentBlock[] | undefined
-  const partial: string[] = []
-  for (const event of events) {
-    const content = assistantMessageOutput(event)
-    if (content !== undefined) {
-      message = content
-    } else if (event.type === 'assistant/chunk' && event.data.chunk.type === 'text-delta') {
-      partial.push(event.data.chunk.text)
-    }
-  }
-  if (message !== undefined) return message
-  const text = partial.join('')
-  return text.length > 0 ? [{ type: 'text', text }] : undefined
+  const fold = new AssistantOutputFold()
+  for (const event of events) fold.push(event)
+  return fold.collect()
 }
--- a/packages/subagent/subagent/src/index.ts
+++ b/packages/subagent/subagent/src/index.ts
@@ -69,7 +69,7 @@ import { snapshotSubagentDescriptor } from './descriptor.ts'
 import { subagentIdentityProjectionDefinition, subagentTimingProjectionDefinition } from './projection.ts'

 export * from './out-of-process.ts'
-export { assistantMessageOutput, finalAssistantOutput } from './assistant-output.ts'
+export { AssistantOutputFold, finalAssistantOutput } from './assistant-output.ts'
 export { SubagentRunId } from './types.ts'
 export type {
  ContinuableCreateRequest,
--- a/packages/subagent/subagent/src/lifecycle.ts
+++ b/packages/subagent/subagent/src/lifecycle.ts
@@ -129,7 +129,9 @@ export function observeRun(
      emit('subagent/end', {
        ...identity,
        stopReason: result.stopReason,
-        lastAssistantMessage: result.output,
+        // One encoding for "no output" across both lifecycle shapes: the
+        // field is absent, matching the continuable epoch edge.
+        ...result.output.length === 0 ? {} : { lastAssistantMessage: result.output },
      }, parent)
    },
    () => {
--- a/packages/subagent/subagent/tests/assistant-output.spec.ts
+++ b/packages/subagent/subagent/tests/assistant-output.spec.ts
@@ -1,7 +1,7 @@
 import { describe, expect, it } from 'vitest'
 import type { ContentBlock } from '@deepseek-ai/dsh-llm'
 import type { SessionEvent } from '@deepseek-ai/dsh-session'
-import { assistantMessageOutput, finalAssistantOutput } from '../src/assistant-output.ts'
+import { AssistantOutputFold, finalAssistantOutput } from '../src/assistant-output.ts'

 function message(content: ContentBlock[]): SessionEvent {
  return { type: 'assistant/message', data: { message: { content } } } as SessionEvent
@@ -15,15 +15,6 @@ function reasoningDelta(text: string): SessionEvent {
  return { type: 'assistant/chunk', data: { chunk: { type: 'reasoning-delta', text } } } as SessionEvent
 }

-describe('assistantMessageOutput', () => {
-  it('returns content only for a non-empty assistant message', () => {
-    const content: ContentBlock[] = [{ type: 'text', text: 'answer' }]
-    expect(assistantMessageOutput(message(content))).toBe(content)
-    expect(assistantMessageOutput(message([]))).toBeUndefined()
-    expect(assistantMessageOutput(textDelta('chunk'))).toBeUndefined()
-  })
-})
-
 describe('finalAssistantOutput', () => {
  it('selects the last non-empty message past a later empty usage-only message', () => {
    const events = [
@@ -58,3 +49,17 @@ describe('finalAssistantOutput', () => {
    expect(finalAssistantOutput([reasoningDelta('thinking'), message([])])).toBeUndefined()
  })
 })
+
+describe('AssistantOutputFold', () => {
+  it('folds raw text pieces into the same streamed fallback (ACP chunk transport)', () => {
+    const fold = new AssistantOutputFold()
+    fold.pushText('partial ')
+    fold.pushText('')
+    fold.pushText('answer')
+    expect(fold.collect()).toEqual([{ type: 'text', text: 'partial answer' }])
+  })
+
+  it('collects undefined until any output is folded', () => {
+    expect(new AssistantOutputFold().collect()).toBeUndefined()
+  })
+})
--- a/packages/subagent/subagent/tests/service.spec.ts
+++ b/packages/subagent/subagent/tests/service.spec.ts
@@ -15,6 +15,7 @@ import SubagentService, {
  type SubagentProvider,
  type SubagentResult,
  type SubagentRun,
+  type SubagentRunEndInfo,
  type SubagentStartRequest,
 } from '@deepseek-ai/dsh-subagent'
 import { SessionId, type SessionEvent } from '@deepseek-ai/dsh-session'
@@ -263,6 +264,17 @@ describe('SubagentService', () => {
      stopReason: 'completed',
    }))

+    // "No output" has ONE encoding on the end edge: the field is absent,
+    // never an empty array, matching the continuable epoch edge.
+    const silent = new StubProvider('silent', NO_CAPS, { output: [], stopReason: 'completed' })
+    subagents.registerProvider(silent)
+    const silentRun = await subagents.start('silent', baseRequest())
+    await silentRun.result
+    await Promise.resolve()
+    const silentEnd = ended.mock.calls.map(call => call[0] as SubagentRunEndInfo).find(info => info.provider === 'silent')
+    expect(silentEnd).toBeDefined()
+    expect('lastAssistantMessage' in silentEnd!).toBe(false)
+
    const failure = Promise.withResolvers<SubagentResult>()
    subagents.registerProvider({
      name: 'infra',
--- a/packages/subagent/tool-subagent/README.i18n.yaml
+++ b/packages/subagent/tool-subagent/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/subagent/tool-subagent/README.md
-README.md: 6ec313b3b97f0ffa7488025d4314b1c6231a6f6a
-README.zh.md: 1fd88363b3ade9d57c194580f81295eed139ac50
+README.md: ac3ec0563cce9128608ca31860b034a103dc1a3a
+README.zh.md: d64831d7cf64800ad3307ce6cb7f294500a0a6f0
--- a/packages/subagent/tool-subagent/README.md
+++ b/packages/subagent/tool-subagent/README.md
@@ -8,7 +8,7 @@ The model-facing delegation tool over one configured `ctx.subagents` provider. C

 Each plugin instance binds one `provider` to one `toolName`; the model receives no provider selector. Load another distinctly named instance to expose another transport. The tool registers only while its provider exists, avoiding sibling load-order and provider-reload dependencies. Its description follows `provider.inheritsParentContext`: fresh children require standalone prompts, while forked children already see completed parent turns.

-A foreground call passes the execution signal through startup and execution, awaits `run.result`, and always awaits `run.dispose()` before returning. Only `completed` returns the canonical `{ kind: 'foreground', runId, output: JsonValue[] }`, rendered as the same final text; abort, refusal, token limit, and other failures become errored tool results without partial output. If result collection and disposal both reject, the errored result preserves both diagnostics.
+A foreground call passes the execution signal through startup and execution, awaits `run.result`, and always awaits `run.dispose()` before returning. Only `completed` returns the canonical `{ kind: 'foreground', runId, output: JsonValue[] }`, rendered as the same final text; abort, refusal, token limit, and other failures become errored tool results whose message appends the child's preserved partial text (the `SubagentResult.output` selection) after the stop-reason headline, so a truncated answer is never reported as success yet never silently lost. If result collection and disposal both reject, the errored result preserves both diagnostics.

 With `run_in_background: true`, `backgroundMode` selects the route. `one-shot` registers a plain parent-owned Task and returns canonical `{ kind: 'background', taskId }`, rendered as `started background subagent task <id>`, even when the provider supports continuable children; generic task tools own its later status, collection, cancellation, and notices. `continuable` requires a provider with the `prepareContinuable` capability, calls `ctx.subagents.startContinuable()`, and returns `{ kind: 'continuable', subagentId }`, rendered as `started subagent <childId>`. The continuable route resolves at inbox acceptance: the child owns its own turns from there, so this call neither waits for nor collects a result, and the child does not report back — its transcript by that id is the source of its output, and the optional global `send_message` tool sends it more work. Starting continuable work does not require `send_message` to be loaded. See the [background subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md), the [continuable subagents Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md), and the [merged-service Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md).

--- a/packages/subagent/tool-subagent/README.zh.md
+++ b/packages/subagent/tool-subagent/README.zh.md
@@ -8,7 +8,7 @@

 每个插件实例把一个 `provider` 绑定到一个 `toolName`；模型不会收到提供方选择器。如需公开另一种传输，请加载另一个名称不同的实例。工具只在其提供方存在时注册，从而避免对同级加载顺序和提供方重新加载的依赖。工具描述遵循 `provider.inheritsParentContext`：新建子 agent（智能体）需要独立提示词，而 fork 子 agent 已能看到父级已完成轮次。

-前台调用会让执行信号贯穿启动和执行，等待 `run.result`，并且在返回前总会等待 `run.dispose()`。只有 `completed` 会返回规范值 `{ kind: 'foreground', runId, output: JsonValue[] }`，并渲染为相同的最终文本；中止、拒绝、token 上限和其他失败都会变成出错的工具结果，不包含局部输出。如果结果收集与 dispose（资源释放）都 reject，出错的结果会保留两项诊断信息。
+前台调用会让执行信号贯穿启动和执行，等待 `run.result`，并且在返回前总会等待 `run.dispose()`。只有 `completed` 会返回规范值 `{ kind: 'foreground', runId, output: JsonValue[] }`，并渲染为相同的最终文本；中止、拒绝、token 上限和其他失败都会变成出错的工具结果，其消息在终止原因标题之后附带子代理保留下来的部分文本（即 `SubagentResult.output` 的选取结果）——被截断的回答不会被报告为成功，也绝不会被悄悄丢弃。如果结果收集与 dispose（资源释放）都 reject，出错的结果会保留两项诊断信息。

 设置 `run_in_background: true` 后，`backgroundMode` 会选择路由。`one-shot` 会注册一个归父级所有的普通 Task，并返回规范值 `{ kind: 'background', taskId }`，渲染为 `started background subagent task <id>`，即使提供方支持可继续子 agent 也不例外；通用 Task 工具负责其后续状态、收集、取消和通知。`continuable` 要求提供方具备 `prepareContinuable` 能力，调用 `ctx.subagents.startContinuable()`，并返回 `{ kind: 'continuable', subagentId }`，渲染为 `started subagent <childId>`。可继续路由在 inbox 接受时结算：子 agent 自此拥有自己的轮次，因此该调用既不等待也不收集结果，而且子 agent 不会回报——通过该 id 查看其 transcript（文本记录）即是其输出来源，可选的全局 `send_message` 工具则向其发送更多工作。启动可继续工作不要求加载 `send_message`。见 [后台 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md)、[可继续的 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md)和[服务合并 Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md)。

--- a/packages/subagent/tool-subagent/src/index.ts
+++ b/packages/subagent/tool-subagent/src/index.ts
@@ -134,6 +134,21 @@ function stopReasonError(result: SubagentResult): string | undefined {
  }
 }

+/**
+ * Append the child's preserved partial answer to a stop-reason error so a
+ * truncated or cancelled child's real text still reaches the parent model.
+ * @param error - the stop-reason headline.
+ * @param output - the child's selected output (`SubagentResult.output`).
+ * @returns the headline, extended with the partial text when any exists.
+ */
+function withPartialText(error: string, output: ContentBlock[]): string {
+  const text = output
+    .filter((block): block is Extract<ContentBlock, { type: 'text' }> => block.type === 'text')
+    .map(block => block.text)
+    .join('')
+  return text.length === 0 ? error : `${error}\nPartial output before the run ended:\n${text}`
+}
+
 type ForegroundToolResult = {
  readonly kind: 'foreground'
  readonly runId: SubagentRun['id']
@@ -149,8 +164,9 @@ async function settleForegroundRun(run: SubagentRun): Promise<ForegroundToolResu
    run.result.then((result): ForegroundToolResult => {
      const error = stopReasonError(result)
      if (error !== undefined) {
-        // The registry converts this throw to isError; partial output is not success.
-        throw new Error(error)
+        // The registry converts this throw to isError; partial output is not
+        // success, but the preserved partial answer still reaches the parent.
+        throw new Error(withPartialText(error, result.output))
      }
      return {
        kind: 'foreground',
--- a/packages/subagent/tool-subagent/tests/tool-subagent.spec.ts
+++ b/packages/subagent/tool-subagent/tests/tool-subagent.spec.ts
@@ -154,6 +154,9 @@ describe('dsh-tool-subagent', () => {
    const result = await callSubagent(ctx, { description: 'd', prompt: 'p' })
    expect(result.isError).toBe(true)
    expect(text(result)).toContain(fragment)
+    // The failure is not partial success, but the child's preserved partial
+    // answer still reaches the parent model inside the error result.
+    expect(text(result)).toContain('scripted subagent reply')
  })

  it('registers under a configurable toolName so multiple providers can coexist', async () => {