dsh-plugin-guide 0.3.2 → 0.3.4

package/references/official-docs/docs/subsystems/llm-streaming.md

@@ -23,12 +23,13 @@ interface ContentBlockMap {
   'text': TextBlock
   'reasoning': ReasoningBlock
   'image': ImageBlock
+  'file': FileBlock
   'tool-call': ToolCallBlock
   'tool-result': ToolResultBlock
 }
 ```
 
-The block interfaces (full fields in source): `TextBlock` (`text`), `ReasoningBlock` (thinking, distinct from visible text), `ImageBlock` (a durable [image attachment](attachment.md)), `ToolCallBlock` (`id: ToolCallId`, `name`, raw-JSON `arguments`), and `ToolResultBlock` (`toolCallId`, nested `content: ContentBlock[]`, `isError?`). `ContentBlock = ContentBlockMap[ContentBlockType]`. A new modality belongs in the merge-extensible map only when its adapter, UI, compaction, and durable replay paths honor it.
+The block interfaces (full fields in source): `TextBlock` (`text`), `ReasoningBlock` (thinking, distinct from visible text), `ImageBlock` (a durable [image attachment](attachment.md)), `FileBlock` (a durable verbatim [file attachment](attachment.md) that request assembly projects to handle text for every route), `ToolCallBlock` (`id: ToolCallId`, `name`, raw-JSON `arguments`), and `ToolResultBlock` (`toolCallId`, nested `content: ContentBlock[]`, `isError?`). `ContentBlock = ContentBlockMap[ContentBlockType]`. A new modality belongs in the merge-extensible map only when its adapter, UI, compaction, and durable replay paths honor it.
 
 Image access belongs to request serialization rather than the durable attachment or deterministic request-image version. `resolveImageAttachmentAccess()` combines the attachment provider's optional host object path with a mapping supplied by the consumer for the current tool execution filesystem. The result is available only for that request and does not participate in `variantId`.
 
@@ -216,6 +217,16 @@ type StreamChunk =
   }
 ```
 
+<a id="compact-assistant-streams"></a>
+
+## Compact Assistant streams
+
+`AssistantStreamAccumulator` pairs each `StreamChunk` with its original safe-integer timestamp and produces `AssistantStreamRecord[]`. Consecutive text, reasoning, or tool-argument deltas for the same block become one record with `time0`, exact timestamp gaps, and one array entry per original delta; every other chunk stays a timestamped raw record. This representation removes repeated event envelopes without joining token boundaries or dropping terminal, usage, block, failure, or replay facts.
+
+`snapshot()` returns a detached immutable stream. `expandAssistantStream()` strictly checks record keys, member counts, indexes, timestamps, tool-call identity, and lossless JSON before recreating the exact timed chunk sequence. The Session log embeds this stream in `assistant/message` for a surface result or `assistant/attempt` for an attempt with no surface message.
+
+Process-local `agent/assistant-stream` frames carry live presentation. Durable replay, telemetry, token accounting, and historical UI assembly expand the embedded settlement instead of treating live frames as persisted facts.
+
 ## `LlmFailure`
 
 Every thrown or in-band final-adapter failure normalizes to one serializable provider-neutral payload. `providerRetryAfterMs` is a validated positive delay requested by the provider, not a retry decision; `ProviderRequestId` is an opaque branded string for diagnostics.
@@ -280,7 +291,7 @@ Every adapter MUST obey these, and every consumer may rely on them:
 
 - **`usage` before `finish`, nothing after `finish`.** Defer both to the provider's end-of-stream marker so a trailing usage-only chunk can't violate the ordering.
 - **Tool-call `arguments` stay raw JSON strings end-to-end.** Partial fragments stream via `argumentsDelta`; a provider that hands back parsed objects re-stringifies at `block-end`.
-- **Two sanctioned error paths, one `LlmFailure` type.** A failure may either THROW from `stream()` (transport/protocol errors) **or** end the stream with `finish {kind:'error'|'aborted', failure}` (provider in-band errors, for adapters that can't throw mid-stream). `LlmError.failure` carries the same `LlmFailure`. After the call selects its adapter, the stream preserves the exact thrown `Error` object and associates immutable facts plus the serving registration's immutable retry policy with that call; the agent loop closes the failed step and offers the error, facts, immutable prior-retried facts, serving policy, and turn signal to `agent/request-error`. A handling listener returns `{ kind: 'retry' }` after its awaited repair; absent recovery the structured failure becomes the turn error, and no normal assistant message or tool side effect is committed for that attempt.
+- **Two sanctioned error paths, one `LlmFailure` type.** A failure may either THROW from `stream()` (transport/protocol errors) **or** end the stream with `finish {kind:'error'|'aborted', failure}` (provider in-band errors, for adapters that can't throw mid-stream). `LlmError.failure` carries the same `LlmFailure`. After the call selects its adapter, the stream preserves the exact thrown `Error` object and associates immutable facts plus the serving registration's immutable retry policy with that call; the agent loop commits the attempt stream as `assistant/attempt`, closes the failed step, and offers the error, facts, immutable prior-retried facts, serving policy, and turn signal to `agent/request-error`. A handling listener returns `{ kind: 'retry' }` after its awaited repair; absent recovery the structured failure becomes the turn error, and no surface Assistant message or tool side effect is committed for that attempt.
 - **One adapter call is one provider attempt.** Adapters disable library retries. Agent-level recovery opens another durable numbered turn; direct `ctx.llm.stream()` callers remain single-attempt.
 - **Provider stalls are bounded at the transport.** Both shipping remote adapters expose positive finite `streamIdleTimeoutMs` with a five-minute default. The watchdog arms only while iterator `next()` is outstanding, uses one stable signal for the whole request, maps its own expiry to `TIMEOUT`, and keeps an earlier caller abort as `ABORTED`.
 - **Context overflow has one canonical code.** Both DeepSeek adapters classify explicit provider detail through `isContextWindowExceededError()` and surface `CONTEXT_WINDOW_EXCEEDED`, whether the failure arrives as a thrown HTTP `LlmError` or an in-band finish error. Consumers route on the code, never provider text.
@@ -951,6 +962,14 @@ providerRetryPolicy(provider: string): ResolvedRetryPolicy
  */
 imageRequestPricing(provider: string, model: string): LlmImageRequestPricing | undefined
 
+/**
+ * Resolve the exact text one durable file occurrence contributes to every
+ * provider request in the current execution environment.
+ * @param ref - durable verbatim file reference from model history.
+ * @returns the same deterministic handle text used at adapter dispatch.
+ */
+fileRequestText(ref: FileAttachmentRef): string
+
 /**
  * Discover models advertised by one registered provider. Catalog membership
  * is advisory and never changes routing or request validation.
@@ -1006,6 +1025,8 @@ async prepareCall(config: LlmCallConfig, signal?: AbortSignal): Promise<Prepared
 stream(options: GenerateOptions): AsyncIterable<StreamChunk>
 ```
 
+Types: [FileAttachmentRef](attachment.md)
+
 Source: [`packages/llm/llm/src/index.ts`](../../packages/llm/llm/src/index.ts)
 
 <a id="llm-events"></a>

package/references/official-docs/docs/subsystems/llm-streaming.zh.md

@@ -23,12 +23,13 @@ interface ContentBlockMap {
   'text': TextBlock
   'reasoning': ReasoningBlock
   'image': ImageBlock
+  'file': FileBlock
   'tool-call': ToolCallBlock
   'tool-result': ToolResultBlock
 }
 ```
 
-各块接口（完整字段见源码）：`TextBlock`（`text`）、`ReasoningBlock`（thinking，区别于可见文本）、`ImageBlock`（一个持久的[图片附件](attachment.zh.md)）、`ToolCallBlock`（`id: ToolCallId`、`name`、原始 JSON `arguments`），以及 `ToolResultBlock`（`toolCallId`、嵌套 `content: ContentBlock[]`、`isError?`）。`ContentBlock = ContentBlockMap[ContentBlockType]`。仅当适配器、UI、压缩（compaction）和持久回放路径均支持某种新模态时，才将其纳入可合并扩展的 map。
+各块接口（完整字段见源码）：`TextBlock`（`text`）、`ReasoningBlock`（thinking，区别于可见文本）、`ImageBlock`（一个持久的[图片附件](attachment.zh.md)）、`FileBlock`（一个持久的原样[文件附件](attachment.zh.md)，请求组装对每条路由都把它投影为 handle 文本）、`ToolCallBlock`（`id: ToolCallId`、`name`、原始 JSON `arguments`），以及 `ToolResultBlock`（`toolCallId`、嵌套 `content: ContentBlock[]`、`isError?`）。`ContentBlock = ContentBlockMap[ContentBlockType]`。仅当适配器、UI、压缩（compaction）和持久回放路径均支持某种新模态时，才将其纳入可合并扩展的 map。
 
 图片访问方式属于请求序列化，不属于持久附件或确定性请求图片版本。`resolveImageAttachmentAccess()` 把附件提供方可选的宿主对象路径，与消费方为当前工具执行文件系统提供的映射组合起来。结果只适用于本次请求，不参与 `variantId`。
 
@@ -216,6 +217,16 @@ type StreamChunk =
   }
 ```
 
+<a id="compact-assistant-streams"></a>
+
+## 紧凑 Assistant stream
+
+`AssistantStreamAccumulator` 把每个 `StreamChunk` 与其原始安全整数时间戳配对，并生成 `AssistantStreamRecord[]`。同一 block 的连续 text、reasoning 或 tool argument delta 会变成一个 record，使用 `time0`、精确时间戳间隔和每个原始 delta 对应的一个数组成员；其他 chunk 保留为带时间戳的 raw record。该表示会移除重复 event envelope，但不会合并 token 边界，也不会丢弃 terminal、usage、block、failure 或 replay 事实。
+
+`snapshot()` 返回分离且不可变的 stream。`expandAssistantStream()` 会严格检查 record key、成员数、index、时间戳、tool-call identity 与无损 JSON，再重建精确的带时间 chunk 序列。Session 日志会把该 stream 嵌入作为 surface result 的 `assistant/message`，或嵌入没有 surface message 的 `assistant/attempt`。
+
+进程本地 `agent/assistant-stream` frame 承载实时呈现。持久回放、遥测、token 记账与历史 UI 组装会展开嵌入式 settlement，而不会把 live frame 当作持久事实。
+
 <a id="llmfailure"></a>
 
 ## `LlmFailure`
@@ -282,7 +293,7 @@ interface LlmImageRequestPricing {
 
 - **`usage` 在 `finish` 之前，`finish` 之后不再有任何分片。** 将两者都推迟到提供方的流结束标记，这样尾部的 usage-only 分片就不会违反顺序。
 - **工具调用的 `arguments` 全程保持原始 JSON 字符串。** 部分片段通过 `argumentsDelta` 流式传输；如果提供方返回的是已解析的对象，适配器在 `block-end` 时重新序列化为字符串。
-- **两条受支持的错误路径，共用一个 `LlmFailure` 类型。** 失败可以从 `stream()` 抛出（传输／协议错误），**或者**以 `finish {kind:'error'|'aborted', failure}` 结束流（无法在流中途抛异常的适配器用它表示提供方带内错误）。`LlmError.failure` 携带同一个 `LlmFailure`。调用选定适配器后，流会保留被抛出的确切 `Error` 对象，并将不可变事实以及实际服务注册所对应的不可变重试策略关联到该调用；agent loop（智能体循环）关闭失败步骤，再把错误、事实、不可变的先前已重试失败事实、实际服务策略和轮次信号提供给 `agent/request-error`。处理该错误的 listener 在其 await 的修复完成后返回 `{ kind: 'retry' }`；若未恢复，结构化失败会成为轮次错误，并且该次尝试不会提交正常 assistant 消息或工具副作用。
+- **两条受支持的错误路径，共用一个 `LlmFailure` 类型。** 失败可以从 `stream()` 抛出（传输／协议错误），**或者**以 `finish {kind:'error'|'aborted', failure}` 结束流（无法在流中途抛异常的适配器用它表示提供方带内错误）。`LlmError.failure` 携带同一个 `LlmFailure`。调用选定适配器后，流会保留被抛出的确切 `Error` 对象，并将不可变事实以及实际服务注册所对应的不可变重试策略关联到该调用；agent loop（智能体循环）先把 attempt stream 提交为 `assistant/attempt`，再关闭失败步骤，并把错误、事实、不可变的先前已重试失败事实、实际服务策略和轮次信号提供给 `agent/request-error`。处理该错误的 listener 在其 await 的修复完成后返回 `{ kind: 'retry' }`；若未恢复，结构化失败会成为轮次错误，并且该次 attempt 不会提交 surface Assistant message 或工具副作用。
 - **一次适配器调用就是一次提供方尝试。** 适配器禁用库重试。agent 层恢复会打开另一个持久、带编号的轮次；直接调用 `ctx.llm.stream()` 的调用方仍然只尝试一次。
 - **提供方停顿在传输层受到时限约束。** 两个已交付的远程适配器都暴露正数且有限的 `streamIdleTimeoutMs`，默认五分钟。watchdog 只在 iterator `next()` 尚未完成时启动，整个请求使用同一个稳定 signal，把自身到期映射为 `TIMEOUT`，并把更早发生的调用方中止保留为 `ABORTED`。
 - **上下文溢出只有一个规范 code。** 两个 DeepSeek 适配器都通过 `isContextWindowExceededError()` 对提供方的显式细节分类并暴露 `CONTEXT_WINDOW_EXCEEDED`，无论失败以抛出的 HTTP `LlmError` 还是带内 finish error 到达。消费方按 code 路由，绝不依赖提供方文本。
@@ -957,6 +968,14 @@ providerRetryPolicy(provider: string): ResolvedRetryPolicy
  */
 imageRequestPricing(provider: string, model: string): LlmImageRequestPricing | undefined
 
+/**
+ * Resolve the exact text one durable file occurrence contributes to every
+ * provider request in the current execution environment.
+ * @param ref - durable verbatim file reference from model history.
+ * @returns the same deterministic handle text used at adapter dispatch.
+ */
+fileRequestText(ref: FileAttachmentRef): string
+
 /**
  * Discover models advertised by one registered provider. Catalog membership
  * is advisory and never changes routing or request validation.
@@ -1012,6 +1031,8 @@ async prepareCall(config: LlmCallConfig, signal?: AbortSignal): Promise<Prepared
 stream(options: GenerateOptions): AsyncIterable<StreamChunk>
 ```
 
+Types: [FileAttachmentRef](attachment.zh.md)
+
 Source: [`packages/llm/llm/src/index.ts`](../../packages/llm/llm/src/index.ts)
 
 <a id="llm-events"></a>

package/references/official-docs/docs/subsystems/persistence.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 docs/subsystems/persistence.md
-persistence.md: cd7eaa126ea16b17526223498d1fa1591792747c
-persistence.zh.md: c96aa87f738e5607c699dfa97e6d2b237767b76d
+persistence.md: 7cce80f43591610c1e0667d480971dbd8f3cd4f7
+persistence.zh.md: 73169ec2f3a58f06a7355922b0e0f7e4e82944c7