billion-context-dsh 0.2.9 → 0.2.11

package/README.en.md

@@ -3,7 +3,7 @@
 [English](./README.en.md) | [中文](./README.md)
 
 > **⚠️ Beta notice — not for production use**
-> This project (**v0.2.9**) is a work-in-progress beta. The [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) itself is also in **public beta**. **Do not use either in engineering / production environments** — expect breaking changes and rough edges.
+> This project (**v0.2.11**) is a work-in-progress beta. The [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) itself is also in **public beta**. **Do not use either in engineering / production environments** — expect breaking changes and rough edges.
 
 <p align="center">
 <strong>Built with gratitude on top of these projects</strong> — please give them a ⭐:
@@ -130,7 +130,7 @@ DSH derives every model request from its append-only session log (the *surface*)
 | nudge ("efficiency note — compress early and keep context lean") | injected at `agent/pre-step` by the kernel's pressure decision — efficiency note + context breakdown + compression rules, tone aligned with kernel/pi; never an order |
 | `decompress` | read-only recovery of shadowed originals from the log |
 | `search_context` | scores a unified doc set (block summaries + shadowed originals) rebuilt from the log via acp-kernel `searchBlocks` (hybrid: stemming + CJK bigrams + char n-gram fuzzy); hits link back to the owning block |
-| `acp_status` | CONTEXT BREAKDOWN (tool/text/summaries shares of the visible total) + compressed-block ledger + nudge decision line; no context-window rows; scope/view/tool/sort/limit drilldown supported |
+| `acp_status` | CONTEXT BREAKDOWN (tool/text/summaries shares of the visible total) + compressed-block ledger + nudge decision line + a `Checkpoint seqs` row mapping each ACTIVE block's kernel ref (`bN`) to its checkpoint summary seq — compressing a checkpoint seq distills that block (issue #60); no context-window rows; scope/view/tool/sort/limit drilldown supported |
 | block state | in-memory kernel state + **log-rebuilt ledger** (no sidecar files) |
 | tiered distillation (T2/T3) | re-compressing a block's summary node distills that block (tier 2); distilling a tier-2 block yields tier 3. Tier + kernel block ids are persisted to the log, so kernel state rehydrates from the log after a restart and stays distillable |
 | compression accounting (shadow price) | `shadowedTokenCount` (what the host occupancy display deducts) is priced in the **host token-meter's vocabulary** (`ctx.tokenMeter.measure` preferred; exact mirror in `src/host-tokens.ts` as fallback) — never the plugin's internal CJK-aware estimate (that is display currency; mixing it into the host ledger can drive `messageTokens` negative and brick a CJK-heavy session, issue #54) |
@@ -150,7 +150,7 @@ A walkthrough of the ACP philosophy this project inherits — how active context
 | `compress` | Replace a seq range with a dense summary you write (edges auto-balanced to tool-pair boundaries); re-compressing a block's summary node distills it (tier 2/3) |
 | `decompress` | Restore a previously compressed block's original content (read-only); accepts the `bN` ref shown by acp_status or a compaction id |
 | `search_context` | Search compressed block summaries and originals by keyword (acp-kernel hybrid retrieval: stemming + CJK bigrams + fuzzy); hits link back to the owning block |
-| `acp_status` | CONTEXT BREAKDOWN (tool/text/summaries shares of the visible total) + compressed-block ledger + nudge decision line; no context-window rows. Drilldown supported: `scope:"compressed"` per block, `scope:"uncompressed"` + `view:"messages"`/`"ranges"` per message/range, with `tool` filter, `sort` order and `limit` cap. Drilldown row refs are kernel ids (mN) — feed them straight to `compress` as `startSeq`/`endSeq` (auto-mapped to the live surface seq); `Surface:` seqs work too |
+| `acp_status` | CONTEXT BREAKDOWN (tool/text/summaries shares of the visible total) + compressed-block ledger + nudge decision line + a `Checkpoint seqs` row mapping each ACTIVE block's kernel ref (`bN`) to its checkpoint summary seq (the distill entry point, issue #60); no context-window rows. Drilldown supported: `scope:"compressed"` per block, `scope:"uncompressed"` + `view:"messages"`/`"ranges"` per message/range, with `tool` filter, `sort` order and `limit` cap. Drilldown row refs are kernel ids (mN) — feed them straight to `compress` as `startSeq`/`endSeq` (auto-mapped to the live surface seq); `Surface:` seqs work too |
 | `/acp` | status / compress / decompress from the command bar; status also shows human-side window info (estimated context, window source, compressed-block ledger, and **nudge arbitration** — `nudge: idle/ACTIVE — reason` plus how many tokens remain until the next nudge, decided by the same kernel turn as the nudge path) |
 
 ## Upstream & credits
@@ -171,7 +171,7 @@ This project reuses `acp-kernel`'s compression core and `billion-context-pi`'s d
 | Key | Default | Meaning |
 |---|---|---|
 | `modelContextLimit` | auto-detected (fallback `128000`) | Context window used for the kernel's pressure decisions; an explicit value wins and skips the probe |
-| `autoModelContextLimit` | `true` | Probe the model's real window from the model API (`agent.ctx.llm.resolveModelInfo`); fall back to the default on failure, the `/acp` command shows the window source (the `acp_status` model tool carries no window info) |
+| `autoModelContextLimit` | `true` | Probe the model's real window from the model API (`agent.ctx.llm.resolveModelInfo`); fall back to the default on failure, the `/acp` command shows the window source (the `acp_status` model tool carries no window info). A failed probe is surfaced in the host log and the `/acp` panel (`restart to re-probe`) — the failure is cached like a success, so fixing the gateway requires a restart or an explicit `modelContextLimit` before the probe retries |
 | `nudgeMinContextLimitPct` | kernel default `0.45` | Nudge window lower bound (usage fraction) — validation only; the growth-driven trigger has no percentage floor — same default as billion-context-pi |
 | `nudgeMaxContextLimitPct` | engine default `0.70` (kernel/pi default `0.75`) | Over-limit line: above this the nudge fires regardless of growth — deliberately below the host compaction-basic 80% auto-compaction line so the forced nudge fires first; an explicit value wins |
 | `nudgeEmergencyThresholdPct` | engine default `0.85` (kernel/pi default `0.95`) | Emergency nudge (bypasses the per-turn dedup) — lowered from `0.95`: at 95% the model has no room to act and the 80% auto-compaction line shadows it; an explicit value wins |

package/README.md

@@ -3,7 +3,7 @@
 [中文](./README.md) | [English](./README.en.md)
 
 > **⚠️ 测试版声明——请勿用于生产环境**
-> 本项目（**v0.2.9**）仍处于开发中的测试版。[DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) 本身也处于**公开测试版**阶段。**请勿将两者用于工程化 / 生产环境**——预期会有破坏性变更与粗糙之处。
+> 本项目（**v0.2.11**）仍处于开发中的测试版。[DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) 本身也处于**公开测试版**阶段。**请勿将两者用于工程化 / 生产环境**——预期会有破坏性变更与粗糙之处。
 
 <p align="center">
 <strong>衷心感谢以下项目——请给它们一个 ⭐：</strong>
@@ -131,7 +131,7 @@ DSH 的每个模型请求都派生自其 append-only 会话日志（*surface*）
 | nudge（"效率提示——尽早压缩保持精简"） | 由内核的压力决策在 `agent/pre-step` 注入——效率通知 + 上下文分解 + 压缩规则，语气对齐 kernel/pi；绝非命令 |
 | `decompress` | 从日志只读恢复被遮蔽的原文 |
 | `search_context` | 从日志重建块摘要 + 被遮蔽原文的统一文档集，交 acp-kernel `searchBlocks`（hybrid：词干化 + CJK bigram + 字符 n-gram 模糊）打分；命中回链所属块 |
-| `acp_status` | CONTEXT BREAKDOWN（tool/text/summaries 占可见总量）+ 压缩块账本 + nudge 决策行；不含上下文窗口；支持 scope/view/tool/sort/limit 钻取 |
+| `acp_status` | CONTEXT BREAKDOWN（tool/text/summaries 占可见总量）+ 压缩块账本 + nudge 决策行 + `Checkpoint seqs` 行（active 块的 `bN → seq` 映射——压缩某个 checkpoint seq 即蒸馏该块，issue #60）；不含上下文窗口；支持 scope/view/tool/sort/limit 钻取 |
 | 块状态 | 内存内核状态 + **日志重建账本**（无旁车文件） |
 | 分层蒸馏（T2/T3） | 再次压缩某块的摘要节点 = 蒸馏该块（tier 2），蒸馏 tier-2 块得 tier 3；tier 与内核块 id 持久化进日志，重启后内核状态从日志再水合、可继续蒸馏 |
 | 压缩记账（影子价格） | `shadowedTokenCount`（宿主占用率据此扣减）**用宿主 token-meter 的词汇计价**（`ctx.tokenMeter.measure` 优先，`src/host-tokens.ts` 精确镜像兜底）——绝不混用插件内部的 CJK 感知估算（那是展示货币，混用会把宿主账本扣成负数、卡死中文会话，issue #54） |
@@ -151,7 +151,7 @@ DSH 的每个模型请求都派生自其 append-only 会话日志（*surface*）
 | `compress` | 用你书写的紧凑摘要替换 seq 范围（边界自动平衡到 tool-call/result 配对点）；对某块的摘要节点再次压缩 = 分层蒸馏（tier 2/3） |
 | `decompress` | 恢复已压缩块的原始内容（只读）；接受 acp_status 显示的 `bN` 或 compaction id |
 | `search_context` | 按关键词搜索压缩块摘要与原文（acp-kernel hybrid 检索：词干化 + CJK bigram + 模糊）；命中回链所属块 |
-| `acp_status` | CONTEXT BREAKDOWN（tool/text/summaries 占可见总量）+ 压缩块账本 + nudge 决策行；不含上下文窗口。支持钻取：`scope:"compressed"` 逐块、`scope:"uncompressed"` + `view:"messages"`/`"ranges"` 逐消息/区间，`tool` 过滤、`sort` 排序、`limit` 截断。钻取行 ref 是内核 mN——可直接作为 `compress` 的 `startSeq`/`endSeq`（自动映射为 live surface seq）；`Surface:` 的 seq 同样可用 |
+| `acp_status` | CONTEXT BREAKDOWN（tool/text/summaries 占可见总量）+ 压缩块账本 + nudge 决策行 + `Checkpoint seqs` 行（active 块的 `bN → seq` 蒸馏入口，issue #60）；不含上下文窗口。支持钻取：`scope:"compressed"` 逐块、`scope:"uncompressed"` + `view:"messages"`/`"ranges"` 逐消息/区间，`tool` 过滤、`sort` 排序、`limit` 截断。钻取行 ref 是内核 mN——可直接作为 `compress` 的 `startSeq`/`endSeq`（自动映射为 live surface seq）；`Surface:` 的 seq 同样可用 |
 | `/acp` | 从命令栏执行 status / compress / decompress；status 额外展示 human-side 窗口信息（estimated context、context window 来源、压缩账本、**nudge 仲裁**——`nudge: idle/ACTIVE — reason` 及距下一次 nudge 还差多少 token，与 nudge 路径同一内核判定） |
 
 ## 上游项目与致谢
@@ -172,7 +172,7 @@ DSH 的每个模型请求都派生自其 append-only 会话日志（*surface*）
 | 键 | 默认值 | 含义 |
 |---|---|---|
 | `modelContextLimit` | 自动探测（回退 `128000`） | 用于内核压力决策的上下文窗口；显式配置时优先且跳过探测 |
-| `autoModelContextLimit` | `true` | 从模型 API 自动探测真实窗口（`agent.ctx.llm.resolveModelInfo`）；探测失败回退默认值，`/acp` 命令展示窗口来源（模型工具 `acp_status` 不含窗口信息） |
+| `autoModelContextLimit` | `true` | 从模型 API 自动探测真实窗口（`agent.ctx.llm.resolveModelInfo`）；探测失败回退默认值，`/acp` 命令展示窗口来源（模型工具 `acp_status` 不含窗口信息）。探测失败会在宿主日志与 `/acp` 面板提示（`restart to re-probe`）——失败结果同样被缓存，修复网关后需重启或显式设置 `modelContextLimit` 才会重新探测 |
 | `nudgeMinContextLimitPct` | 内核默认 `0.45` | Nudge 窗口下界（用量占比）——仅作配置校验，增长路径的触发没有百分比下限——与 billion-context-pi 相同的默认值 |
 | `nudgeMaxContextLimitPct` | engine 默认 `0.70`（内核/pi 默认 `0.75`） | 过限线：超过此值则无论增长与否都触发 nudge——刻意低于宿主 compaction-basic 的 80% 自动压缩线，保证强制 nudge 先触发；显式配置优先 |
 | `nudgeEmergencyThresholdPct` | engine 默认 `0.85`（内核/pi 默认 `0.95`） | 紧急 nudge（绕过每轮去重）——从 `0.95` 下调：95% 时模型已无操作空间且会被 80% 自动压缩线遮蔽；显式配置优先 |

package/dist/commands.d.ts

@@ -4,6 +4,6 @@
  * @module billion-context-dsh/commands
  */
 import type { CommandDefinition } from '@deepseek-ai/dsh-commands';
-import type { ToolEnvironment } from './tools.ts';
+import { type ToolEnvironment } from './tools.ts';
 /** Register the /acp command (idempotent per engine). */
 export declare function acpCommand(env: ToolEnvironment): CommandDefinition;

package/dist/index.js

@@ -3525,7 +3525,7 @@ var NUDGE_ALLOWED = {
   normal: /* @__PURE__ */ new Set(["pct", "philosophy"]),
   emergency: /* @__PURE__ */ new Set(["pct", "philosophy"]),
   guidance: /* @__PURE__ */ new Set(),
-  tier: /* @__PURE__ */ new Set(["tier", "count", "prevTier", "tokens", "seqs"]),
+  tier: /* @__PURE__ */ new Set(["tier", "count", "prevTier", "tokens", "seqs", "firstSeq", "lastSeq"]),
   breakdown: /* @__PURE__ */ new Set(["system", "tool", "summaries", "code", "text"]),
   growth: /* @__PURE__ */ new Set(["growth"]),
   tip: /* @__PURE__ */ new Set()
@@ -3600,7 +3600,7 @@ var DEFAULT_PROMPTS = {
     normal: "This is an efficiency nudge to compress early and keep context lean \u2014 not an overflow warning. A separate, stronger alert will appear if the context is actually full.\n\n{philosophy}",
     emergency: "\u26A0\uFE0F Context limit reached \u2014 compress now. Prioritize consumed tool outputs.\n\n{philosophy}",
     guidance: HOW_TO_COMPRESS_RULES,
-    tier: "Tier {tier}: {count} tier-{prevTier} block(s) distillable ({tokens} tokens) \u2014 compress their summary node(s) [seqs {seqs}] to reclaim the original messages.",
+    tier: "Tier {tier}: {count} tier-{prevTier} block(s) distillable ({tokens} tokens) \u2014 distill them by compressing their checkpoint seq(s) [seqs {seqs}] as one range: compress({ content: [{ startSeq: {firstSeq}, endSeq: {lastSeq}, summary }] }).",
     breakdown: "Context breakdown: {system}K system | {tool}K tool | {summaries}K summaries | {code}K code | {text}K text",
     growth: "+{growth}K since last nudge",
     tip: "\u{1F4A1} Compress all ranges in one call (pass multiple content entries: `content: [{...}, {...}]`)."
@@ -3749,7 +3749,7 @@ function replaceTierTrigger(text, nudge, session, prompts) {
   const next = rest.match(/\n\nHOW TO COMPRESS/);
   const end = next !== null ? start + 2 + next.index : text.length;
   const targets = nudge.tierTargetBlocks;
-  const summarySeqs = targets.map((block) => summarySeqOfKernelBlock(session, block.blockId)).filter((seq) => seq !== null);
+  const summarySeqs = targets.map((block) => summarySeqOfKernelBlock(session, block.blockId)).filter((seq) => seq !== null).sort((a, b) => a - b);
   const pending = nudge.tier === 2 ? nudge.breakdown?.pendingT2 : nudge.breakdown?.pendingT3;
   const tokens = typeof pending === "number" ? pending : 0;
   const tierValue = nudge.tier === null ? 2 : nudge.tier;
@@ -3758,7 +3758,9 @@ function replaceTierTrigger(text, nudge, session, prompts) {
     count: targets.length,
     prevTier: tierValue - 1,
     tokens,
-    seqs: summarySeqs.join(", ")
+    seqs: summarySeqs.join(", "),
+    firstSeq: summarySeqs[0] ?? "n/a",
+    lastSeq: summarySeqs[summarySeqs.length - 1] ?? "n/a"
   });
   return text.slice(0, start) + "\n\n" + tierLine + text.slice(end);
 }
@@ -3795,7 +3797,7 @@ function renderNudgeFromTemplates(nudge, emergency, session, prompts) {
   if (prompts.nudge.guidance !== "") parts.push("", prompts.nudge.guidance);
   if ((nudge.tier === 2 || nudge.tier === 3) && (nudge.tierTargetBlocks?.length ?? 0) > 0) {
     const targets = nudge.tierTargetBlocks;
-    const summarySeqs = targets.map((block) => summarySeqOfKernelBlock(session, block.blockId)).filter((seq) => seq !== null);
+    const summarySeqs = targets.map((block) => summarySeqOfKernelBlock(session, block.blockId)).filter((seq) => seq !== null).sort((a, b) => a - b);
     const pending = nudge.tier === 2 ? nudge.breakdown?.pendingT2 : nudge.breakdown?.pendingT3;
     const tokens = typeof pending === "number" ? pending : 0;
     const tierLine = renderTemplate(prompts.nudge.tier, {
@@ -3803,7 +3805,9 @@ function renderNudgeFromTemplates(nudge, emergency, session, prompts) {
       count: targets.length,
       prevTier: nudge.tier - 1,
       tokens,
-      seqs: summarySeqs.join(", ")
+      seqs: summarySeqs.join(", "),
+      firstSeq: summarySeqs[0] ?? "n/a",
+      lastSeq: summarySeqs[summarySeqs.length - 1] ?? "n/a"
     });
     if (tierLine !== "") parts.push(tierLine);
     const tierRules = nudge.tier === 2 ? TIER2_DISTILL_RULES : TIER3_CONDENSE_RULES;
@@ -3832,6 +3836,9 @@ function requireAgent(exec) {
   }
   return exec.agent;
 }
+async function resolveEffectiveWindow(env, agent) {
+  return env.windowFor === void 0 ? { limit: env.modelContextLimit, source: "explicit" } : await env.windowFor(agent);
+}
 var compressParameters = {
   // Tolerated wrapped-arguments form: some models emit
   // `{ "arguments": "{\"content\": [...]}" }` (double-nested) or
@@ -3943,7 +3950,8 @@ async function handleCompress(env, args, exec) {
   const coreMessages = allLogMessages(session);
   const surfaceMessages = eventsToCoreMessages(surfaceEventsOf(session));
   const tokenCount = resolveTokenCount(agent, surfaceMessages);
-  const config = kernelConfigFor(env);
+  const window = await resolveEffectiveWindow(env, agent);
+  const config = kernelConfigFor({ ...env, modelContextLimit: window.limit });
   const turn = env.kernel.processTurn({ messages: coreMessages, state, config, tokenCount });
   env.store.set(session, turn.state);
   const byRaw = turn.state.messageRefs.byRaw;
@@ -4068,7 +4076,7 @@ async function handleCompress(env, args, exec) {
       effectiveMessageIds: block.effectiveMessageIds
     });
     const adjusted = start !== range.startSeq || end !== range.endSeq;
-    const tierLabel2 = tier === 1 ? "" : `, tier ${tier}`;
+    const tierLabel2 = `, tier ${tier}`;
     const note = range.recovered === true ? ` (seqs ${range.startSeq}..${range.endSeq} were already shadowed \u2014 compressed the live remainder ${start}..${end})` : adjusted ? ` (adjusted from ${range.startSeq}..${range.endSeq} to balanced edges)` : "";
     lines.push(
       `  block ${compactionId.slice(0, 8)}: seqs ${start}..${end}, ${shadowed.length} messages shadowed${tierLabel2}${note}`
@@ -4227,7 +4235,8 @@ async function handleStatus(env, rawArgs, exec) {
   const coreMessages = allLogMessages(session);
   const surfaceMessages = eventsToCoreMessages(surface, toolNames);
   const tokenCount = resolveTokenCount(agent, surfaceMessages);
-  const config = kernelConfigFor(env);
+  const window = await resolveEffectiveWindow(env, agent);
+  const config = kernelConfigFor({ ...env, modelContextLimit: window.limit });
   const turn = env.kernel.processTurn({ messages: coreMessages, state, config, tokenCount });
   const statusMessages = eventsToCoreMessages(
     surface.filter((event) => !isCheckpointEvent(event)),
@@ -4240,6 +4249,10 @@ async function handleStatus(env, rawArgs, exec) {
     if (nudge !== void 0) {
       lines.push("", `Nudge: ${nudge.shouldInject ? "ACTIVE" : "idle"} \u2014 ${nudge.reason}`);
     }
+    const checkpointRows = blockRegistry(session).filter((entry) => entry.active && entry.summarySeq !== null).map((entry) => `${entry.kernelBlockId} \u2192 seq ${entry.summarySeq}`);
+    if (checkpointRows.length > 0) {
+      lines.push("", `Checkpoint seqs (active blocks \u2014 compress a checkpoint seq to distill it): ${checkpointRows.join(", ")}`);
+    }
   }
   lines.push("", `Surface: ${surfaceSummary(session)}`);
   if (args.scope === "uncompressed") {
@@ -4296,6 +4309,7 @@ function windowSourceLabel(window) {
   if (window.source === "auto") {
     return `auto-detected from ${window.provider ?? "?"}/${window.model ?? "?"}`;
   }
+  if (window.probeFailed === true) return "default (auto-detection failed \u2014 restart to re-probe)";
   return "default (auto-detection unavailable)";
 }
 async function detectContextWindow(agent, provider, model) {
@@ -4319,7 +4333,7 @@ async function statusText(env, agent) {
   const coreMessages = allLogMessages(session);
   const surfaceMessages = eventsToCoreMessages(surfaceEventsOf(session));
   const estimated = resolveTokenCount(agent, surfaceMessages);
-  const window = env.windowFor === void 0 ? { limit: env.modelContextLimit, source: "explicit" } : await env.windowFor(agent);
+  const window = await resolveEffectiveWindow(env, agent);
   const limit = window.limit;
   const lines = [
     `ACP status \u2014 session ${session.id}`,
@@ -4328,6 +4342,9 @@ async function statusText(env, agent) {
     `  estimated context: ${estimated} / ${limit} (${Math.round(estimated / limit * 100)}%)`,
     `  context window: ${limit} (${windowSourceLabel(window)})`
   ];
+  if (window.probeFailed === true) {
+    lines.push(`  \u26A0 window auto-detection failed \u2014 using the ${limit} fallback (restart to re-probe, or set modelContextLimit explicitly)`);
+  }
   const state = structuredClone(env.store.stateFor(session));
   const config = kernelConfigFor({ ...env, modelContextLimit: limit });
   const turn = env.kernel.processTurn({ messages: coreMessages, state, config, tokenCount: estimated });
@@ -4561,7 +4578,14 @@ var AcpCompactionEngine = class extends CompactionEngine {
       window = { limit: DEFAULT_CONTEXT_WINDOW, source: "default", provider, model };
     } else {
       const detected = await detectContextWindow(agent, provider, model);
-      window = detected === null ? { limit: DEFAULT_CONTEXT_WINDOW, source: "default", provider, model } : { limit: detected, source: "auto", provider, model };
+      if (detected === null) {
+        this.ctx.logger.warn(
+          `billion-context-dsh: context-window auto-detection failed for ${provider}/${model} \u2014 using the ${DEFAULT_CONTEXT_WINDOW} fallback (restart to re-probe, or set modelContextLimit explicitly)`
+        );
+        window = { limit: DEFAULT_CONTEXT_WINDOW, source: "default", provider, model, probeFailed: true };
+      } else {
+        window = { limit: detected, source: "auto", provider, model };
+      }
     }
     this.windowCache.set(key, window);
     return window;