@modusensus/dsh-mneme 0.7.20 → 0.7.21

package/README.en.md

@@ -166,6 +166,7 @@ Every **background LLM call** (autoDream consolidation + summary, autoSummarize
 
 | Version | Highlights |
 |------|------|
+| **v0.7.21** | Fixed autoDream/sleep effort fallback being dead code on the stream path (the catch-based retry from v0.7.16 never fired): dsh-llm rc.1 turns adapter-stage failures (incl. `UNSUPPORTED_REASONING_EFFORT`) into a terminal error finish chunk instead of a throw; `streamText` now captures the finish-chunk failure cause (`describeStreamFailure` normalizes `{code,message}`) + `withEffortFallback` gains a `getStreamError` accessor (retries without effort when rejected) + `runAuditedLlm` supports `spec.streamError` (audit `error_message` carries the real cause; `run.error` stays a stable `"llm failed"`); 688 tests green |
 | **v0.7.20** | Heat model restored (issue #87): v0.7.0 self-evolving memory back (`src/heat.js` power-law decay `H=1/(1+λΔt)^α` + per-type half-lives), sleep demotion dual protection (cold time-window + heat<0.05 + importance<5), touchRecalled gating back on `heatEnabled`, entity heat projection (ego node heat → size/opacity), recall_runs recording on by default; **default OFF** (matches v0.7.12 behavior out of the box) + feature_flags whitelist rollback switch + lightMode linkage + sleep demotion audit counts exposed (workbench can show "N demoted") + phase-two frontend (/list heat projection, HeatBadge three-tier badge, self-gated status heat-distribution card, order=heat page-local sort, all self-gated); better-sidebar fix (issue #88: soft integration moved to an inner dynamic sub-plugin, no more startup failure without bs); 685 tests green |
 | **v0.7.18** | Ecosystem step one + query convergence: better-sidebar soft integration (inject declaration + optional peer `dsh-better-sidebar` + registerTab reusing the four views, safe skip when absent; narrow-container `@container` adaptation) + `/list?deposited=only` view (receipt_chain ∪ source=dream) + deposited/archived filter chips in the library + status-page dashboard (server totals + view-all jumps) + drawer restore for archived memories; 667 tests green |
 | **v0.7.17** | Panel polish: sidebar entry continuously aligned with the host (MutationObserver syncs the New-Session class + `width:100%` + native centering restored, resilient to async skin rewrites) + importance rendered as Lucide star glyphs (`ImportanceStars` filled/hollow component replacing text ★) + toolbar dropdown stacking fix (z-index moved to the container; export/import menu no longer painted under the sticky month header); 664 tests green |
@@ -212,7 +213,12 @@ Every **background LLM call** (autoDream consolidation + summary, autoSummarize
 | v0.7.1–0.7.8 | ✅ Done | Issue fixes + graph backfill | tags↔entity_attrs bridge, inline-confirm delete, sidebar trigger toggle, brace escaping, user/fact layered types + stats endpoint (later removed), prefix-id resolution (later removed), sleep batch entity extraction, snapshotEvents() DSH compat; 764→812 tests green |
 | v0.7.9 | ✅ Done | lib-sync gate | Issue #65: src-only shim silently killed the shipped lib — synced lib + pre-publish src↔lib consistency check + lib smoke tests; 815 tests green |
 | v0.7.10–0.7.12 | ✅ Done | Panel redesign + near-rewrite | UX upgrades (color dots, pan/zoom, pagination, global search), then near-rewrite: inlined panel → pure HTTP API + standalone CLI + lightMode; TYPES narrowed 8→6 |
-| v0.7.13–0.7.20 | ✅ Done | Distillation, security, desktop, dream fix, panel polish, ecosystem, heat restore | codingRetrospect + 429 governor; private reasoning blocks dropped (CWE-200); desktop panel redesign + feature flags; autoDream thinking-model empty-body fix + backfilled tests; sidebar entry alignment + star glyphs + menu stacking polish; better-sidebar soft integration (inner dynamic sub-plugin fix, issue #88) + deposited/archived library views; heat model restored (issue #87) with phase-two frontend + archival audit (issue #87); 685 tests green |
+| v0.7.13–0.7.15 | ✅ Done | Distillation, security, desktop | codingRetrospect + 429 governor; private reasoning blocks dropped (CWE-200); desktop panel redesign + feature flags + status dashboard + import/export |
+| v0.7.16 | ✅ Done | autoDream thinking-model fix | config-first routing restored (Issue #25) + reasoningEffort auto-retry + honest audit on parse failure; backfilled API-route & lib smoke tests; 662 tests green |
+| v0.7.17 | ✅ Done | Panel polish | Sidebar entry continuously aligned with the host (MutationObserver syncs the New-Session class + `width:100%` + native centering, resilient to async skin rewrites) + Lucide star glyphs (`ImportanceStars`) + toolbar dropdown stacking fix; 664 tests green |
+| v0.7.18 | ✅ Done | Ecosystem step one + query convergence | better-sidebar soft integration (inject declaration + optional peer + registerTab reusing the four views, safe skip when absent; narrow-container `@container`) + `/list?deposited=only` view + deposited/archived filter chips + status-page dashboard + drawer restore; 667 tests green |
+| v0.7.20 | ✅ Done | Heat restore + phase-two frontend + better-sidebar fix | Heat model fully restored (issue #87, backported from v0.7.10: power-law decay + TYPE_DECAY + sleep heat-combined dual protection + entity heat projection) + acceptance checklist landed (heatEnabled default OFF / feature_flags 31 keys / lightMode linkage / sleep demotion audit / updated_at⊥last_accessed_at contract) + phase-two frontend (/list heat projection, HeatBadge three-tier, order=heat page-local sort) + better-sidebar fix (issue #88: inner dynamic sub-plugin); 685 tests green |
+| v0.7.21 | ✅ Done | effort-fallback stream fix | Catch-based effort fallback was dead code on the stream path (dsh-llm rc.1 turns adapter failures into a terminal error finish chunk instead of a throw) → `streamText` captures the finish-chunk cause (`describeStreamFailure`) + `withEffortFallback` gains a `getStreamError` accessor (retries without effort when rejected) + `runAuditedLlm` supports `spec.streamError` (real cause in audit); 688 tests green |
 | **v0.8.0** | 🚧 Planned (late Sep) | Graph enhancement | Interest-drift visualization + scope isolation (issue #17) + cross-workspace sharing |
 
 > All new capabilities ship as **toggleable features** (enabled/disabled via configuration), conservatively on by default and never breaking existing behavior. The `failure_memories` table and the autoDream decision engine have already paved the way for future reflective growth.

package/README.md

@@ -207,6 +207,7 @@ v0.3.0 起新增**记忆基因**层：从记忆里抽取**命名实体**、**带
 
 | 版本 | 亮点 |
 |------|------|
+| **v0.7.21** | 修复 autoDream/sleep 的 effort 回退在流式路径失效（v0.7.16 的 catch 式回退是死代码）：dsh-llm rc.1 把 adapter 阶段异常（含 `UNSUPPORTED_REASONING_EFFORT`）转成终态 error finish chunk 不再抛出；`streamText` 现捕获 finish-chunk 失败原因（新增 `describeStreamFailure` 归一化 `{code,message}`）+ `withEffortFallback` 增加 `getStreamError` 访问器（effort 被拒时去掉重试一次）+ `runAuditedLlm` 支持 `spec.streamError`（audit 行 `error_message` 携带真实原因，`run.error` 稳定 `"llm failed"` 不变）；688 测试全绿 |
 | **v0.7.20** | heat 热度模型回归（issue #87）：找回 v0.7.0 自进化记忆（`src/heat.js` 幂律衰减 `H=1/(1+λΔt)^α` + per-type 差异化半衰期）、sleep 降级热联合双保护（时间窗冷 + heat<0.05 + importance<5）、touchRecalled 门控改回 `heatEnabled`、实体热投影（ego 节点 heat → 前端大小/明暗）、recall_runs 默认记录；**默认关**（默认=与 v0.7.12 行为一致）+ feature_flags 白名单回滚开关 + lightMode 联动 + sleep 降级审计计数暴露（工作动态可展示"降级 N 条"）+ 阶段二前端（/list heat 投影、HeatBadge 三档徽章、状态页热度分布卡、order=heat 页内排序，全部自门控）；better-sidebar 修复（issue #88：软集成改内层动态子插件，无 bs 环境不再启动失败）；685 测试全绿 |
 | **v0.7.18** | 生态第一步 + 查询收敛：better-sidebar 软集成（inject 声明 + optional peer `dsh-better-sidebar` + registerTab 复用四视图，未装安全跳过；窄容器 `@container` 适配）+ `/list?deposited=only` 沉淀视图（receipt_chain ∪ source=dream）+ 记忆库沉淀/已归档筛选 chip + 状态页仪表盘化（统计 + 查看全部跳转预置筛选）+ 抽屉归档记忆「恢复」；667 测试全绿 |
 | **v0.7.17** | 面板体验细化：侧边栏入口持续对齐宿主（MutationObserver 同步「新会话」类名 + `width:100%` + 交还原生居中，皮肤异步改写不再失配）+ 重要性星级换 Lucide 星形（`ImportanceStars` 实心/空心组件，替换文本 ★）+ 工具栏下拉层级修复（z-index 提到容器，导出/导入菜单不再被吸顶月份头遮挡）；664 测试全绿 |
@@ -286,6 +287,10 @@ v0.3.0 起新增**记忆基因**层：从记忆里抽取**命名实体**、**带
 | **v0.7.14** | ✅ 完成 | 安全修复（CWE-200） | 蒸馏不再采集私有推理块：`collectMessages` 只采公开 text，防止记忆库沉淀模型私有思考链；617 测试全绿 |
 | **v0.7.15** | ✅ 完成 | 桌面端适配 | 记忆库面板重设计 + 功能开关 30 键 UI（features API）+ 状态页工作台 + 导入导出（镜像同构 md 黄金闭环）+ Token 面板默认遮蔽；645 测试全绿 |
 | **v0.7.16** | ✅ 完成 | autoDream thinking 模型空体修复 + 补测 | 恢复 config-first 路由（设置面板「巩固模型」生效，Issue #25）+ reasoningEffort 被拒自动去掉重试 + 解析失败如实记 llm_audit error；补测 API 路由空白（/delete、/entities、/external-api）+ lib 运行时冒烟；662 测试全绿 |
+| **v0.7.17** | ✅ 完成 | 面板体验细化 | 侧边栏入口持续对齐宿主（MutationObserver 同步「新会话」类名 + `width:100%` + 交还原生居中，皮肤异步改写不再失配）+ 重要性星级换 Lucide 星形（`ImportanceStars` 实心/空心组件，替换文本 ★）+ 工具栏下拉层级修复（z-index 提到容器，导出/导入菜单不再被吸顶月份头遮挡）；664 测试全绿 |
+| **v0.7.18** | ✅ 完成 | 生态第一步 + 查询收敛 | better-sidebar 软集成（inject 声明 + optional peer `dsh-better-sidebar` + registerTab 复用四视图，未装安全跳过；窄容器 `@container` 适配）+ `/list?deposited=only` 沉淀视图（receipt_chain ∪ source=dream）+ 记忆库沉淀/已归档筛选 chip + 状态页仪表盘化（统计 + 查看全部跳转预置筛选）+ 抽屉归档记忆「恢复」；667 测试全绿 |
+| **v0.7.20** | ✅ 完成 | heat 回归 + 阶段二前端 + better-sidebar 修复 | heat 热度模型完整找回（issue #87，v0.7.10 移植：幂律衰减 + TYPE_DECAY + sleep 热联合双保护 + 实体热投影）+ 验收清单落地（heatEnabled 默认关 / feature_flags 31 键 / lightMode 联动 / sleep 降级审计暴露 / updated_at⊥last_accessed_at 契约）+ 阶段二前端（/list heat 投影、HeatBadge 三档、order=heat 页内排序）+ better-sidebar 修复（issue #88：内层动态子插件）；685 测试全绿 |
+| **v0.7.21** | ✅ 完成 | effort 回退流式修复 | autoDream/sleep 的 catch 式 effort 回退在流式路径是死代码（dsh-llm rc.1 把 adapter 异常转成终态 error finish chunk 不再抛出）→ `streamText` 捕获 finish-chunk 失败原因（`describeStreamFailure` 归一化）+ `withEffortFallback` 增加 `getStreamError` 访问器（effort 被拒去重试）+ `runAuditedLlm` 支持 `spec.streamError`（audit 记真实原因）；688 测试全绿 |
 | **v0.8.0** | 🚧 计划中（9 月末） | 图谱增强 | 兴趣漂移可视化 + scope 隔离（issue #17）+ 跨 workspace 记忆共享 |
 
 > 新能力一律做成**可开关的功能**（配置启用/关闭），默认保守开启、不破坏现有行为。`failure_memories` 表与 autoDream 决策引擎已为后续反思性成长铺好路。

package/lib/dream/sleep.js

@@ -17,7 +17,7 @@
 import { randomUUID, createHash } from "node:crypto";
 import { validateDecisions, applyDecisions } from "./decisions.js";
 import { findPotentialConflicts } from "./clustering.js";
-import { buildReceipt, withEffortFallback } from "../dream.js";
+import { buildReceipt, describeStreamFailure, withEffortFallback } from "../dream.js";
 import { computeHeat } from "../heat.js";
 
 const SUMMARY_MAX = 120;
@@ -61,12 +61,17 @@ function parseJsonArray(text) {
 }
 
 /** Same stream consumption contract as dream.js. */
-async function streamText(ctx, options) {
+async function streamText(ctx, options, onStreamError) {
   if (!ctx?.llm?.stream) return undefined;
   let text = "";
   for await (const chunk of ctx.llm.stream(options)) {
     if (chunk.type === "text-delta" && typeof chunk.text === "string") text += chunk.text;
     if (chunk.type === "finish" && (chunk.reason?.kind === "error" || chunk.reason?.kind === "aborted")) {
+      // Same rc.1 error-as-finish-chunk behavior as dream.js — surface the
+      // cause instead of discarding it.
+      if (typeof onStreamError === "function") {
+        try { onStreamError(chunk.reason); } catch { /* diagnostics only */ }
+      }
       return undefined;
     }
   }
@@ -186,7 +191,10 @@ async function phaseConflicts(ctx, service, config, logger, runId, semantic = nu
     `候选冲突：\nid=${p.a.id} | type=${p.a.type} | title=${p.a.title}\n${p.a.content}\n---\nid=${p.b.id} | type=${p.b.type} | title=${p.b.title}\n${p.b.content}\n（相似度 ${p.similarity.toFixed(2)}）`
   ).join("\n\n");
   const sleepEffort = config.sleepReasoningEffort && config.sleepReasoningEffort !== "none" ? config.sleepReasoningEffort : null;
-  const runConflict = (withEffort) => streamText(ctx, {
+  let conflictStreamFailure = "";
+  const runConflict = (withEffort) => {
+    conflictStreamFailure = "";
+    return streamText(ctx, {
     provider: route.provider,
     model: route.model,
     purpose: "sleep-conflict",
@@ -196,9 +204,13 @@ async function phaseConflicts(ctx, service, config, logger, runId, semantic = nu
       { role: "system", content: [{ type: "text", text: CONFLICT_PROMPT }] },
       { role: "user", content: [{ type: "text", text: listText }] }
     ]
-  });
-  const text = await withEffortFallback(ctx, sleepEffort, () => runConflict(true), () => runConflict(false));
-  if (text === undefined) return { status: "failed", error: "llm failed" };
+  }, (reason) => { conflictStreamFailure = describeStreamFailure(reason); });
+  };
+  const text = await withEffortFallback(ctx, sleepEffort, () => runConflict(true), () => runConflict(false), () => conflictStreamFailure);
+  if (text === undefined) {
+    if (conflictStreamFailure) ctx.logger?.warn?.(`dsh-mneme sleep: conflict stream aborted or errored (${conflictStreamFailure})`);
+    return { status: "failed", error: "llm failed" };
+  }
   const decisions = parseJsonArray(text);
   if (!decisions) return { status: "failed", error: "invalid decisions json" };
   // validateDecisions 要求每个 snapshot id 恰好被 claim 一次。v0.4.4 起它本身
@@ -298,7 +310,10 @@ async function phasePatterns(ctx, service, config, logger, runId, signal = null)
     .join("\n");
   const maxPatterns = config.sleepMaxPatternPerRun ?? 3;
   const sleepEffort = config.sleepReasoningEffort && config.sleepReasoningEffort !== "none" ? config.sleepReasoningEffort : null;
-  const runPattern = (withEffort) => streamText(ctx, {
+  let patternStreamFailure = "";
+  const runPattern = (withEffort) => {
+    patternStreamFailure = "";
+    return streamText(ctx, {
     provider: route.provider,
     model: route.model,
     purpose: "sleep-pattern",
@@ -308,9 +323,13 @@ async function phasePatterns(ctx, service, config, logger, runId, signal = null)
       { role: "system", content: [{ type: "text", text: PATTERN_PROMPT.replace("N", String(maxPatterns)) }] },
       { role: "user", content: [{ type: "text", text: listText }] }
     ]
-  });
-  const text = await withEffortFallback(ctx, sleepEffort, () => runPattern(true), () => runPattern(false));
-  if (text === undefined) return { status: "failed", error: "llm failed" };
+  }, (reason) => { patternStreamFailure = describeStreamFailure(reason); });
+  };
+  const text = await withEffortFallback(ctx, sleepEffort, () => runPattern(true), () => runPattern(false), () => patternStreamFailure);
+  if (text === undefined) {
+    if (patternStreamFailure) ctx.logger?.warn?.(`dsh-mneme sleep: pattern stream aborted or errored (${patternStreamFailure})`);
+    return { status: "failed", error: "llm failed" };
+  }
   const decisions = parseJsonArray(text);
   if (!decisions || decisions.length === 0) return { status: "skipped", reason: "no patterns found" };
   // Evidence ids are provenance refs; an LLM-fabricated id would mint a dead

package/lib/dream.js

@@ -1,7 +1,7 @@
 import { validateDecisions, applyDecisions } from "./dream/decisions.js";
 import { clusterMemories, findPotentialConflicts } from "./dream/clustering.js";
 import { createHash, randomUUID } from "node:crypto";
-export { validateDecisions, applyDecisions, withEffortFallback };
+export { validateDecisions, applyDecisions, withEffortFallback, describeStreamFailure };
 
 
 // Extract the first JSON array from LLM output, tolerating markdown fences,
@@ -238,18 +238,34 @@ function buildRecordReceipts({ runId, committed, snapshot, policyEpoch }) {
  * surfaces as undefined. The caller decides how to treat an empty result.
  * `onUsage` (optional, Bug8) receives any usage chunk for token accounting.
  */
-async function streamText(ctx, options, onUsage) {
+async function streamText(ctx, options, onUsage, onStreamError) {
   let text = "";
   for await (const chunk of ctx.llm.stream(options)) {
     if (chunk.type === "text-delta" && typeof chunk.text === "string") text += chunk.text;
     if (chunk.type === "usage" && typeof onUsage === "function") onUsage(chunk);
     if (chunk.type === "finish" && (chunk.reason?.kind === "error" || chunk.reason?.kind === "aborted")) {
+      // dsh-llm rc.1 turns adapter-stage failures (unknown provider route,
+      // UNSUPPORTED_REASONING_EFFORT from resolveCallWithInfo, …) into a
+      // terminal finish chunk instead of a throw — the cause rides in
+      // chunk.reason.failure {message, code}. Surface it, never swallow it.
+      if (typeof onStreamError === "function") {
+        try { onStreamError(chunk.reason); } catch { /* diagnostics only */ }
+      }
       return undefined;
     }
   }
   return text;
 }
 
+/** One-line human-readable cause from a finish-chunk failure reason. */
+function describeStreamFailure(reason) {
+  const failure = reason?.failure ?? reason ?? {};
+  const code = failure.code ? String(failure.code) : "";
+  const message = String(failure.message ?? failure.error ?? "");
+  if (code && message) return message.includes(code) ? message : `${code}: ${message}`;
+  return code || message;
+}
+
 /**
  * Bug8: wrap a background LLM call so its token/time/status are recorded in the
  * llm_audit_logs table. Best-effort bookkeeping: a failure to WRITE the audit
@@ -279,9 +295,12 @@ async function runAuditedLlm(ctx, service, config, spec, body) {
     });
     if (result === undefined) {
       // stream aborted/errored: the caller treats undefined as a failed run;
-      // record it as error here so the audit shows the truth.
+      // record it as error here so the audit shows the truth. spec.streamError
+      // (a getter) lets the caller attach the finish-chunk cause so the audit
+      // row names it instead of a bare "aborted".
       status = "error";
-      errorMessage = errorMessage ?? "llm stream aborted or errored";
+      const streamErr = typeof spec.streamError === "function" ? String(spec.streamError() ?? "") : "";
+      errorMessage = errorMessage ?? (streamErr ? `llm stream aborted or errored (${streamErr})` : "llm stream aborted or errored");
     } else if (typeof spec.auditError === "function") {
       // A stream that returned text but yields nothing usable is still a
       // failed call — record it as error, not the default success, so the
@@ -328,10 +347,22 @@ async function runAuditedLlm(ctx, service, config, spec, body) {
  * accepted → reasoning capped; rejected → provider default (old behavior),
  * logged so the rejection is observable.
  */
-async function withEffortFallback(ctx, effort, attempt, fallback) {
+async function withEffortFallback(ctx, effort, attempt, fallback, getStreamError) {
   if (!effort || effort === "none") return attempt();
   try {
-    return await attempt();
+    const result = await attempt();
+    if (result === undefined) {
+      // dsh-llm rc.1 streams a provider effort-rejection as a terminal error
+      // finish chunk (adapterStream catches everything, never throws) — match
+      // on the chunk's failure reason here or the retry below is dead code
+      // for the stream path.
+      const reason = String(getStreamError?.() ?? "");
+      if (/reasoning[\s_]*effort|UNSUPPORTED_REASONING_EFFORT/i.test(reason)) {
+        ctx.logger?.warn?.(`dsh-mneme dream: reasoningEffort "${effort}" rejected via stream (${reason}); retrying without it`);
+        return fallback();
+      }
+    }
+    return result;
   } catch (error) {
     const message = String(error?.message ?? error);
     // matches both "reasoning effort" (natural language) and the bare
@@ -630,11 +661,15 @@ export function createDreamScheduler({ onRun, thresholdCount = 10, thresholdChar
     // 记 audit error 并在日志带原始输出前 300 字节，便于定位"推理吞预算返回空体"。
     const effort = config.dreamReasoningEffort && config.dreamReasoningEffort !== "none" ? config.dreamReasoningEffort : null;
     let decisions = null;
-    const runConsolidation = (withEffort) => runAuditedLlm(ctx, service, config, {
+    let streamFailure = "";
+    const runConsolidation = (withEffort) => {
+      streamFailure = "";
+      return runAuditedLlm(ctx, service, config, {
       triggerSource: "autoDream",
       operationType: "dream_consolidate",
       modelId: `${route.provider}:${route.model}`,
       relatedMemoryIds: [...snapshot.keys()],
+      streamError: () => streamFailure,
       auditError: (text) => {
         decisions = extractJsonArray(text);
         return Array.isArray(decisions) ? null : "no json array in llm output";
@@ -649,18 +684,19 @@ export function createDreamScheduler({ onRun, thresholdCount = 10, thresholdChar
         { role: "system", content: [{ type: "text", text: consolidationPrompt }] },
         { role: "user", content: [{ type: "text", text: listText }] }
       ]
-    }, reportUsage));
+    }, reportUsage, (reason) => { streamFailure = describeStreamFailure(reason); }));
+    };
     try {
       // Bug8: the consolidation call is audited (tokens/time/status). A throw
       // re-propagates to the catch below; an aborted stream returns undefined
       // and is treated as a failed run after the check below.
-      decisionText = await withEffortFallback(ctx, effort, () => runConsolidation(true), () => runConsolidation(false));
+      decisionText = await withEffortFallback(ctx, effort, () => runConsolidation(true), () => runConsolidation(false), () => streamFailure);
     } catch (error) {
       logger?.warn?.(`dsh-mneme dream: consolidation llm call failed: ${String(error)}`);
       return finish({ ok: false, error: "llm failed", summary: false });
     }
     if (decisionText === undefined) {
-      logger?.warn?.("dsh-mneme dream: consolidation llm stream aborted or errored");
+      logger?.warn?.(`dsh-mneme dream: consolidation llm stream aborted or errored${streamFailure ? ` (${streamFailure})` : ""}`);
       return finish({ ok: false, error: "llm failed", summary: false });
     }
     if (!Array.isArray(decisions)) {
@@ -782,11 +818,15 @@ export function createDreamScheduler({ onRun, thresholdCount = 10, thresholdChar
     // Summary generation (second LLM call). A throwing stream is reported as
     // a failed run; summary:false marks a run that produced no summary.
     let summaryText;
-    const runSummary = (withEffort) => runAuditedLlm(ctx, service, config, {
+    let summaryStreamFailure = "";
+    const runSummary = (withEffort) => {
+      summaryStreamFailure = "";
+      return runAuditedLlm(ctx, service, config, {
       triggerSource: "autoDream",
       operationType: "dream_summarize",
       modelId: `${route.provider}:${route.model}`,
-      relatedMemoryIds: []
+      relatedMemoryIds: [],
+      streamError: () => summaryStreamFailure
     }, (reportUsage) => streamText(ctx, {
       provider: route.provider,
       model: route.model,
@@ -797,10 +837,11 @@ export function createDreamScheduler({ onRun, thresholdCount = 10, thresholdChar
         { role: "system", content: [{ type: "text", text: SUMMARY_PROMPT }] },
         { role: "user", content: [{ type: "text", text: service.all().filter((m) => !m.archived && m.type !== "summary").map((m) => `- ${m.title}: ${m.content}`).join("\n") }] }
       ]
-    }, reportUsage));
+    }, reportUsage, (reason) => { summaryStreamFailure = describeStreamFailure(reason); }));
+    };
     try {
       // Bug8: the summary call is audited too (operation dream_summarize).
-      summaryText = await withEffortFallback(ctx, effort, () => runSummary(true), () => runSummary(false));
+      summaryText = await withEffortFallback(ctx, effort, () => runSummary(true), () => runSummary(false), () => summaryStreamFailure);
     } catch (error) {
       logger?.warn?.(`dsh-mneme dream: summary llm call failed: ${String(error)}`);
       return finish({ ok: false, error: "llm failed", applied, decisions: auditDecisions, outcome, frozen: frozenCount, summary: false });

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@modusensus/dsh-mneme",
   "description": "Cross-session memory plugin for DeepSeek Harness with autoDream consolidation: SQLite store, Markdown mirrors, 7 model tools, automatic injection, session summarization, user profile/rules, custom slash commands, vector (semantic) search, and a Web GUI panel",
-  "version": "0.7.20",
+  "version": "0.7.21",
   "license": "MIT",
   "repository": {
     "type": "git",

package/src/dream/sleep.js

@@ -17,7 +17,7 @@
 import { randomUUID, createHash } from "node:crypto";
 import { validateDecisions, applyDecisions } from "./decisions.js";
 import { findPotentialConflicts } from "./clustering.js";
-import { buildReceipt, withEffortFallback } from "../dream.js";
+import { buildReceipt, describeStreamFailure, withEffortFallback } from "../dream.js";
 import { computeHeat } from "../heat.js";
 
 const SUMMARY_MAX = 120;
@@ -61,12 +61,17 @@ function parseJsonArray(text) {
 }
 
 /** Same stream consumption contract as dream.js. */
-async function streamText(ctx, options) {
+async function streamText(ctx, options, onStreamError) {
   if (!ctx?.llm?.stream) return undefined;
   let text = "";
   for await (const chunk of ctx.llm.stream(options)) {
     if (chunk.type === "text-delta" && typeof chunk.text === "string") text += chunk.text;
     if (chunk.type === "finish" && (chunk.reason?.kind === "error" || chunk.reason?.kind === "aborted")) {
+      // Same rc.1 error-as-finish-chunk behavior as dream.js — surface the
+      // cause instead of discarding it.
+      if (typeof onStreamError === "function") {
+        try { onStreamError(chunk.reason); } catch { /* diagnostics only */ }
+      }
       return undefined;
     }
   }
@@ -186,7 +191,10 @@ async function phaseConflicts(ctx, service, config, logger, runId, semantic = nu
     `候选冲突：\nid=${p.a.id} | type=${p.a.type} | title=${p.a.title}\n${p.a.content}\n---\nid=${p.b.id} | type=${p.b.type} | title=${p.b.title}\n${p.b.content}\n（相似度 ${p.similarity.toFixed(2)}）`
   ).join("\n\n");
   const sleepEffort = config.sleepReasoningEffort && config.sleepReasoningEffort !== "none" ? config.sleepReasoningEffort : null;
-  const runConflict = (withEffort) => streamText(ctx, {
+  let conflictStreamFailure = "";
+  const runConflict = (withEffort) => {
+    conflictStreamFailure = "";
+    return streamText(ctx, {
     provider: route.provider,
     model: route.model,
     purpose: "sleep-conflict",
@@ -196,9 +204,13 @@ async function phaseConflicts(ctx, service, config, logger, runId, semantic = nu
       { role: "system", content: [{ type: "text", text: CONFLICT_PROMPT }] },
       { role: "user", content: [{ type: "text", text: listText }] }
     ]
-  });
-  const text = await withEffortFallback(ctx, sleepEffort, () => runConflict(true), () => runConflict(false));
-  if (text === undefined) return { status: "failed", error: "llm failed" };
+  }, (reason) => { conflictStreamFailure = describeStreamFailure(reason); });
+  };
+  const text = await withEffortFallback(ctx, sleepEffort, () => runConflict(true), () => runConflict(false), () => conflictStreamFailure);
+  if (text === undefined) {
+    if (conflictStreamFailure) ctx.logger?.warn?.(`dsh-mneme sleep: conflict stream aborted or errored (${conflictStreamFailure})`);
+    return { status: "failed", error: "llm failed" };
+  }
   const decisions = parseJsonArray(text);
   if (!decisions) return { status: "failed", error: "invalid decisions json" };
   // validateDecisions 要求每个 snapshot id 恰好被 claim 一次。v0.4.4 起它本身
@@ -298,7 +310,10 @@ async function phasePatterns(ctx, service, config, logger, runId, signal = null)
     .join("\n");
   const maxPatterns = config.sleepMaxPatternPerRun ?? 3;
   const sleepEffort = config.sleepReasoningEffort && config.sleepReasoningEffort !== "none" ? config.sleepReasoningEffort : null;
-  const runPattern = (withEffort) => streamText(ctx, {
+  let patternStreamFailure = "";
+  const runPattern = (withEffort) => {
+    patternStreamFailure = "";
+    return streamText(ctx, {
     provider: route.provider,
     model: route.model,
     purpose: "sleep-pattern",
@@ -308,9 +323,13 @@ async function phasePatterns(ctx, service, config, logger, runId, signal = null)
       { role: "system", content: [{ type: "text", text: PATTERN_PROMPT.replace("N", String(maxPatterns)) }] },
       { role: "user", content: [{ type: "text", text: listText }] }
     ]
-  });
-  const text = await withEffortFallback(ctx, sleepEffort, () => runPattern(true), () => runPattern(false));
-  if (text === undefined) return { status: "failed", error: "llm failed" };
+  }, (reason) => { patternStreamFailure = describeStreamFailure(reason); });
+  };
+  const text = await withEffortFallback(ctx, sleepEffort, () => runPattern(true), () => runPattern(false), () => patternStreamFailure);
+  if (text === undefined) {
+    if (patternStreamFailure) ctx.logger?.warn?.(`dsh-mneme sleep: pattern stream aborted or errored (${patternStreamFailure})`);
+    return { status: "failed", error: "llm failed" };
+  }
   const decisions = parseJsonArray(text);
   if (!decisions || decisions.length === 0) return { status: "skipped", reason: "no patterns found" };
   // Evidence ids are provenance refs; an LLM-fabricated id would mint a dead

package/src/dream.js

@@ -1,7 +1,7 @@
 import { validateDecisions, applyDecisions } from "./dream/decisions.js";
 import { clusterMemories, findPotentialConflicts } from "./dream/clustering.js";
 import { createHash, randomUUID } from "node:crypto";
-export { validateDecisions, applyDecisions, withEffortFallback };
+export { validateDecisions, applyDecisions, withEffortFallback, describeStreamFailure };
 
 
 // Extract the first JSON array from LLM output, tolerating markdown fences,
@@ -238,18 +238,34 @@ function buildRecordReceipts({ runId, committed, snapshot, policyEpoch }) {
  * surfaces as undefined. The caller decides how to treat an empty result.
  * `onUsage` (optional, Bug8) receives any usage chunk for token accounting.
  */
-async function streamText(ctx, options, onUsage) {
+async function streamText(ctx, options, onUsage, onStreamError) {
   let text = "";
   for await (const chunk of ctx.llm.stream(options)) {
     if (chunk.type === "text-delta" && typeof chunk.text === "string") text += chunk.text;
     if (chunk.type === "usage" && typeof onUsage === "function") onUsage(chunk);
     if (chunk.type === "finish" && (chunk.reason?.kind === "error" || chunk.reason?.kind === "aborted")) {
+      // dsh-llm rc.1 turns adapter-stage failures (unknown provider route,
+      // UNSUPPORTED_REASONING_EFFORT from resolveCallWithInfo, …) into a
+      // terminal finish chunk instead of a throw — the cause rides in
+      // chunk.reason.failure {message, code}. Surface it, never swallow it.
+      if (typeof onStreamError === "function") {
+        try { onStreamError(chunk.reason); } catch { /* diagnostics only */ }
+      }
       return undefined;
     }
   }
   return text;
 }
 
+/** One-line human-readable cause from a finish-chunk failure reason. */
+function describeStreamFailure(reason) {
+  const failure = reason?.failure ?? reason ?? {};
+  const code = failure.code ? String(failure.code) : "";
+  const message = String(failure.message ?? failure.error ?? "");
+  if (code && message) return message.includes(code) ? message : `${code}: ${message}`;
+  return code || message;
+}
+
 /**
  * Bug8: wrap a background LLM call so its token/time/status are recorded in the
  * llm_audit_logs table. Best-effort bookkeeping: a failure to WRITE the audit
@@ -279,9 +295,12 @@ async function runAuditedLlm(ctx, service, config, spec, body) {
     });
     if (result === undefined) {
       // stream aborted/errored: the caller treats undefined as a failed run;
-      // record it as error here so the audit shows the truth.
+      // record it as error here so the audit shows the truth. spec.streamError
+      // (a getter) lets the caller attach the finish-chunk cause so the audit
+      // row names it instead of a bare "aborted".
       status = "error";
-      errorMessage = errorMessage ?? "llm stream aborted or errored";
+      const streamErr = typeof spec.streamError === "function" ? String(spec.streamError() ?? "") : "";
+      errorMessage = errorMessage ?? (streamErr ? `llm stream aborted or errored (${streamErr})` : "llm stream aborted or errored");
     } else if (typeof spec.auditError === "function") {
       // A stream that returned text but yields nothing usable is still a
       // failed call — record it as error, not the default success, so the
@@ -328,10 +347,22 @@ async function runAuditedLlm(ctx, service, config, spec, body) {
  * accepted → reasoning capped; rejected → provider default (old behavior),
  * logged so the rejection is observable.
  */
-async function withEffortFallback(ctx, effort, attempt, fallback) {
+async function withEffortFallback(ctx, effort, attempt, fallback, getStreamError) {
   if (!effort || effort === "none") return attempt();
   try {
-    return await attempt();
+    const result = await attempt();
+    if (result === undefined) {
+      // dsh-llm rc.1 streams a provider effort-rejection as a terminal error
+      // finish chunk (adapterStream catches everything, never throws) — match
+      // on the chunk's failure reason here or the retry below is dead code
+      // for the stream path.
+      const reason = String(getStreamError?.() ?? "");
+      if (/reasoning[\s_]*effort|UNSUPPORTED_REASONING_EFFORT/i.test(reason)) {
+        ctx.logger?.warn?.(`dsh-mneme dream: reasoningEffort "${effort}" rejected via stream (${reason}); retrying without it`);
+        return fallback();
+      }
+    }
+    return result;
   } catch (error) {
     const message = String(error?.message ?? error);
     // matches both "reasoning effort" (natural language) and the bare
@@ -630,11 +661,15 @@ export function createDreamScheduler({ onRun, thresholdCount = 10, thresholdChar
     // 记 audit error 并在日志带原始输出前 300 字节，便于定位"推理吞预算返回空体"。
     const effort = config.dreamReasoningEffort && config.dreamReasoningEffort !== "none" ? config.dreamReasoningEffort : null;
     let decisions = null;
-    const runConsolidation = (withEffort) => runAuditedLlm(ctx, service, config, {
+    let streamFailure = "";
+    const runConsolidation = (withEffort) => {
+      streamFailure = "";
+      return runAuditedLlm(ctx, service, config, {
       triggerSource: "autoDream",
       operationType: "dream_consolidate",
       modelId: `${route.provider}:${route.model}`,
       relatedMemoryIds: [...snapshot.keys()],
+      streamError: () => streamFailure,
       auditError: (text) => {
         decisions = extractJsonArray(text);
         return Array.isArray(decisions) ? null : "no json array in llm output";
@@ -649,18 +684,19 @@ export function createDreamScheduler({ onRun, thresholdCount = 10, thresholdChar
         { role: "system", content: [{ type: "text", text: consolidationPrompt }] },
         { role: "user", content: [{ type: "text", text: listText }] }
       ]
-    }, reportUsage));
+    }, reportUsage, (reason) => { streamFailure = describeStreamFailure(reason); }));
+    };
     try {
       // Bug8: the consolidation call is audited (tokens/time/status). A throw
       // re-propagates to the catch below; an aborted stream returns undefined
       // and is treated as a failed run after the check below.
-      decisionText = await withEffortFallback(ctx, effort, () => runConsolidation(true), () => runConsolidation(false));
+      decisionText = await withEffortFallback(ctx, effort, () => runConsolidation(true), () => runConsolidation(false), () => streamFailure);
     } catch (error) {
       logger?.warn?.(`dsh-mneme dream: consolidation llm call failed: ${String(error)}`);
       return finish({ ok: false, error: "llm failed", summary: false });
     }
     if (decisionText === undefined) {
-      logger?.warn?.("dsh-mneme dream: consolidation llm stream aborted or errored");
+      logger?.warn?.(`dsh-mneme dream: consolidation llm stream aborted or errored${streamFailure ? ` (${streamFailure})` : ""}`);
       return finish({ ok: false, error: "llm failed", summary: false });
     }
     if (!Array.isArray(decisions)) {
@@ -782,11 +818,15 @@ export function createDreamScheduler({ onRun, thresholdCount = 10, thresholdChar
     // Summary generation (second LLM call). A throwing stream is reported as
     // a failed run; summary:false marks a run that produced no summary.
     let summaryText;
-    const runSummary = (withEffort) => runAuditedLlm(ctx, service, config, {
+    let summaryStreamFailure = "";
+    const runSummary = (withEffort) => {
+      summaryStreamFailure = "";
+      return runAuditedLlm(ctx, service, config, {
       triggerSource: "autoDream",
       operationType: "dream_summarize",
       modelId: `${route.provider}:${route.model}`,
-      relatedMemoryIds: []
+      relatedMemoryIds: [],
+      streamError: () => summaryStreamFailure
     }, (reportUsage) => streamText(ctx, {
       provider: route.provider,
       model: route.model,
@@ -797,10 +837,11 @@ export function createDreamScheduler({ onRun, thresholdCount = 10, thresholdChar
         { role: "system", content: [{ type: "text", text: SUMMARY_PROMPT }] },
         { role: "user", content: [{ type: "text", text: service.all().filter((m) => !m.archived && m.type !== "summary").map((m) => `- ${m.title}: ${m.content}`).join("\n") }] }
       ]
-    }, reportUsage));
+    }, reportUsage, (reason) => { summaryStreamFailure = describeStreamFailure(reason); }));
+    };
     try {
       // Bug8: the summary call is audited too (operation dream_summarize).
-      summaryText = await withEffortFallback(ctx, effort, () => runSummary(true), () => runSummary(false));
+      summaryText = await withEffortFallback(ctx, effort, () => runSummary(true), () => runSummary(false), () => summaryStreamFailure);
     } catch (error) {
       logger?.warn?.(`dsh-mneme dream: summary llm call failed: ${String(error)}`);
       return finish({ ok: false, error: "llm failed", applied, decisions: auditDecisions, outcome, frozen: frozenCount, summary: false });

package/test/reasoning-effort.test.js

@@ -234,3 +234,117 @@ test("issue#9: sleep forwards sleepReasoningEffort on its LLM passes", async ()
   }
   store.close();
 });
+
+// ------------------------------------------------------------------ stream-level rejection
+// dsh-llm rc.1 converts adapter-stage failures (including the provider's
+// UNSUPPORTED_REASONING_EFFORT throw from resolveCallWithInfo) into a terminal
+// error finish chunk inside adapterStream — the rejection NEVER reaches our
+// catch. The v0.7.16 throw-based fallback was therefore dead code for the
+// stream path; these tests pin the finish-chunk-based fallback.
+
+test("rc.1 stream-level effort rejection (error finish chunk) also triggers the no-effort retry", async () => {
+  const store = createStore(":memory:");
+  const service = createService({ store, mirror: null, config: {} });
+  const dream = createDreamScheduler({ onRun: () => Promise.resolve({ ok: true, skipped: true }) });
+  const { memory: a } = service.saveWithDedupe({ type: "project", title: "插件", content: "旧", importance: 3 });
+  const { memory: b } = service.saveWithDedupe({ type: "project", title: "插件2", content: "新细节", importance: 4 });
+  const calls = [];
+  const warnings = [];
+  const ctx = {
+    logger: { warn: (m) => warnings.push(String(m)) },
+    agentDefaultModel: { currentSelection: () => ({ provider: "mock", model: "mock-model" }) },
+    llm: {
+      async *stream(options) {
+        calls.push(options);
+        if (options.reasoningEffort) {
+          yield {
+            type: "finish",
+            reason: {
+              kind: "error",
+              failure: {
+                code: "UNSUPPORTED_REASONING_EFFORT",
+                message: 'provider "mock" model "mock-model" does not support reasoning effort "low"'
+              }
+            }
+          };
+          return;
+        }
+        const userText = options.messages.find((m) => m.role === "user")?.content?.[0]?.text ?? "";
+        if (userText.startsWith("id=")) {
+          yield { type: "text-delta", index: 0, text: JSON.stringify([
+            { action: "merge", ids: [a.id, b.id], keepSource: b.id, title: "合并标题", content: "合并内容", importance: 4 }
+          ]) };
+        } else {
+          yield { type: "text-delta", index: 0, text: "记忆库总览：用户偏好中文。" };
+        }
+        yield { type: "finish", reason: { kind: "stop" } };
+      }
+    }
+  };
+  const result = await dream.runDream(ctx, service, { dreamReasoningEffort: "low" });
+  assert.equal(result.ok, true, "run survives the stream-level effort rejection");
+  assert.ok(result.applied > 0, "consolidation still lands changes");
+  assert.equal(calls[0].reasoningEffort, "low", "first attempt forwards the effort");
+  assert.equal("reasoningEffort" in calls[1], false, "retry omits the rejected effort field");
+  assert.ok(warnings.some((w) => w.includes("rejected via stream")), "the stream-level rejection is logged");
+  store.close();
+});
+
+test("non-effort stream failures are not retried and the finish-chunk cause reaches the audit row", async () => {
+  const store = createStore(":memory:");
+  const service = createService({ store, mirror: null, config: {} });
+  service.saveWithDedupe({ type: "project", title: "主题", content: "内容" });
+  const calls = [];
+  const ctx = {
+    logger: { warn: () => {} },
+    agentDefaultModel: { currentSelection: () => ({ provider: "mock", model: "mock-model" }) },
+    llm: {
+      async *stream(options) {
+        calls.push(options);
+        yield { type: "finish", reason: { kind: "error", failure: { code: "PROVIDER_GONE", message: "provider mock is not registered" } } };
+      }
+    }
+  };
+  const dream = createDreamScheduler({ thresholdCount: 1, thresholdChars: 0, delayMs: 0 });
+  const result = await dream.runDream(ctx, service, { dreamReasoningEffort: "low" });
+  assert.equal(result.ok, false);
+  assert.equal(result.error, "llm failed", "the run error stays the stable short string");
+  assert.equal(calls.length, 1, "no blind retry when the stream failure is not an effort rejection");
+  const row = service.listLlmAudits().find((r) => r.operation_type === "dream_consolidate");
+  assert.ok(row && row.status === "error", "failed consolidation still audited");
+  assert.ok(
+    String(row.error_message).includes("PROVIDER_GONE") && String(row.error_message).includes("provider mock is not registered"),
+    "audit error_message carries the finish-chunk cause"
+  );
+  store.close();
+});
+
+test("sleep passes the stream failure accessor so a stream-level effort rejection retries", async () => {
+  const { store, service, vectorIndex } = sleepSetup();
+  const a = service.saveWithDedupe({ type: "project", title: "主题X", content: "内容A 关于主题X", importance: 3 }).memory;
+  const b = service.saveWithDedupe({ type: "project", title: "主题X副本", content: "内容B 关于主题X", importance: 3 }).memory;
+  vectorIndex.saveEmbedding(a.id, [1, 0, 0]);
+  vectorIndex.saveEmbedding(b.id, [1, 0, 0]);
+  const captured = [];
+  const ctx = sleepCtx(null, { provider: "mock", model: "sleep-model" }, captured);
+  ctx.llm.stream = async function* (options) {
+    captured.push(options);
+    if (options.reasoningEffort) {
+      yield {
+        type: "finish",
+        reason: { kind: "error", failure: { code: "UNSUPPORTED_REASONING_EFFORT", message: 'provider "mock" model "sleep-model" does not support reasoning effort "low"' } }
+      };
+      return;
+    }
+    const userText = options.messages.find((m) => m.role === "user")?.content?.[0]?.text ?? "";
+    yield { type: "text-delta", index: 0, text: userText.startsWith("候选冲突")
+      ? JSON.stringify([{ action: "conflict", winner: a.id, loser: b.id, reason: "重复覆盖" }])
+      : "[]" };
+    yield { type: "finish", reason: { kind: "stop" } };
+  };
+  const result = await runSleep(ctx, service, baseConfig({ sleepReasoningEffort: "low" }), ctx.logger, { embedder, vectorIndex }, null);
+  assert.equal(result.status, "ok", "sleep survives the stream-level effort rejection");
+  assert.equal(captured[0].reasoningEffort, "low", "first conflict attempt forwards the effort");
+  assert.equal("reasoningEffort" in captured[1], false, "conflict retry omits the rejected effort field");
+  store.close();
+});