thincoder 0.12.67 → 0.12.69

package/CHANGELOG.md

@@ -1,3 +1,35 @@
+## [0.12.69] — 2026-10-03
+
+> 0.12.68 → 0.12.69（连续号——发布时定号）
+
+### Changed
+
+- **核依赖升级**：`@thincoder/core` `^0.10.1` → `^0.10.2`（统一 provider 解析随核生效）。
+- **无效 provider 态分流（#841）**：TUI——`fallback`（可运行但未设默认模型）⇒ 不弹选择器 + 一行明示（`模型` 缺 ⇒ 词形仅渠道名）；真无效（无 provider ∕ 无 key）⇒ 选择器 + 引导提示行。headless——真无效 ⇒ 可读错误 + 明确退出码；`fallback` ⇒ stderr 一行 + **继续运行**（不退出）。
+
+## [0.12.68] — 2026-10-03
+
+> 0.12.67 → 0.12.68（连续号——发布时定号）
+
+### Added
+
+- **空闲唤醒**：挂起会话可被定时器唤醒——到点自动续跑（可见面 + 中止适配；timer-wake #443 ∕ #444）。
+- **台账健康信号**：台账库异常 ⇒ TUI 警告行（空值以 `—` 占位）。
+- **heap 快照采集运行期开关**（`diagnostics.heapSnapshot`——默认关）。
+- **`PROJECT-MANIFEST.json` 声明面**：codePaths ∕ index ∕ advisor 三键入驻（`conventions.json` 退役）；工程写门豁免 aux 路径（test/tests/scripts、`.thincoder/tmp`）。
+
+### Changed
+
+- **核依赖升级**：`@thincoder/core` `^0.9.5` → `^0.10.1`（公开仓读取 ∥ 会话锚解析 ∥ 多仓机制 ∥ 纪律面随核生效）。
+- **交互整批（parity B1–B10 + 走查轮）**：滚动退出 ∥ 挂起队列 ∥ 状态栏 ∥ 渲染性能 ∥ i18n ∥ 设置 ∥ 复制 ∥ 卡顿指示 ∥ 发送忙碌队列。
+- **测试面重整**：旧套件族退役（runner 骨架保留——空 = 绿）；单测 = 批内件（随批档）。
+
+### Fixed
+
+- **tech-debt 轮 6 ∕ 8**：slash 直通 ∥ 模态门 ∥ sticky bit 重臂 ∥ 级联清除 ∥ aborted reason ∥ advisor 槽守卫 ∥ 台账写者 ∥ timer 支收敛。
+- **批档 CRLF 容忍 + 粘贴生命周期**（#732 ∕ #733）。
+- **会话台账读写守卫**（项目相对守卫面 + 读守卫）。
+
 ## [0.12.67] — 2026-09-25
 
 > 0.12.66 → 0.12.67（连续号——发布时定号）

package/README.md

@@ -17,8 +17,8 @@ resolved from the npm registry for releases).
 - **Checkpoint system**: auto-snapshot before every user task, `list`/`create`/`rewind` tools for the model, single-file restore — rewinding itself is reversible (pre-rewind state auto-saved)
 - **Codebase understanding** ⭐0.5.0: `repo_outline` (dependency outline, auto-injected at startup), `code_search` (source FTS5 + vectors + JSDoc extraction), `doc_search` (docs chunked by ## headings) — background indexing, auto-incremental updates on file writes, three tools
 guided by "structure → intent → details"
-- **Model adaptation** ⭐: top-tier only, latest only. Built-in flagship models from twenty-one providers — DeepSeek / Kimi / Kimi For Coding / GLM / GLM Coding Plan (智谱 Coding) / Qwen / Qwen Token Plan / MiMo (小米) / MiMo Token Plan (小米套餐) / MiniMax / OpenAI / Claude / Gemini /
-Grok / Mistral / Volcengine Ark (豆包) / Hunyuan (腾讯混元) / Tencent TokenHub (腾讯混元网关) / SiliconFlow (硅基流动) / OpenRouter / Groq. No legacy model compatibility. Auto-matched context windows, truncation-resume protocols (prefix/partial), thinking-mode APIs (thinking.type /
+- **Model adaptation** ⭐: top-tier only, latest only. Built-in flagship models from twenty-two providers — DeepSeek / Kimi / Kimi For Coding / GLM / GLM Coding Plan (智谱 Coding) / Qwen / Qwen Token Plan / MiMo (小米) / MiMo Token Plan (小米套餐) / MiniMax / OpenAI / Claude / Gemini /
+Grok / Mistral / Volcengine Ark (豆包) / Hunyuan (腾讯混元) / Tencent TokenHub (腾讯混元网关) / Huawei Cloud MaaS (华为云 MaaS) / SiliconFlow (硅基流动) / OpenRouter / Groq. No legacy model compatibility. Auto-matched context windows, truncation-resume protocols (prefix/partial), thinking-mode APIs (thinking.type /
 reasoning_effort), reasoning_content echo strategies (reasoningEcho), output limits, temperature range clamping — all deeply adapted.
 - **Toolset**: `read` / `write` / `edit` / `bash` / `glob` (supports `**`) / `grep` / `websearch` / `ls` / `fetch` + `read_image` (image/video paste) + three retrieval tools + MCP — all zero-dependency, file tools confined to the working directory
 - **Memory system**: three layers (personal/project/team), FTS5 + vector RRF hybrid retrieval, git-friendly markdown format
@@ -80,6 +80,8 @@ thincoder
 First launch walks you through a setup wizard: arrow keys to pick a provider (built-in presets or a custom endpoint) → enter API key → optionally enter an embedding key (SiliconFlow, enables vector memory search, skippable) → arrow keys to pick a model — no hand-editing config
 files. Adjust anytime with `/model` (add/remove providers, set keys, custom endpoints) or `/config embedkey`. `chat`/`distill` also offer in-place interactive setup when no key is configured in a terminal (in pipes/CI they exit with an error and instructions).
 
+Any OpenAI-compatible endpoint works without a built-in preset: pick **add provider → custom** in `/model` (name + baseURL + model), or hand-write the entry in `providers[]`. Regional endpoint variants work the same way — e.g. Huawei Cloud MaaS `https://api-ap-southeast-1.modelarts-maas.com/openai/v1`.
+
 You can also hand-write `~/.thincoder/config.json` (see "Configuration" below), then:
 
 ```bash
@@ -91,6 +93,7 @@ thincoder memory put --type=rule --title="code style" --content="no semicolons"
 thincoder memory search "code style"
 thincoder memory list
 thincoder memory remove 1
+thincoder memory sweep
 
 # Team memory (optional, available after configuring memory.team)
 thincoder sync                       # pull the team repo and rebuild the index
@@ -110,7 +113,8 @@ encoding/command issues), `/think` (thinking mode toggle and reasoning effort),
 sessions), `/reindex` (rebuild the index), `/extract` (extract knowledge from the current session), `/restore` (restore checkpoint), `/clear`, `/exit`. High-frequency commands support abbreviations: `/h` `/x` `/m` `/p` `/t` `/c` `/n`. Typing `/` shows live matching hints in the
 status bar. Model picker supports search/filter — type to narrow down results.
 
-Configuration comes exclusively from `~/.thincoder/config.json` — no environment-variable configuration is supported.
+Configuration comes exclusively from `~/.thincoder/config.json` — the single source of truth. Sensitive value fields (e.g. a provider's `apiKey` or `websearch.apiKey`)
+may hold a `${env:VAR}` reference, resolved at use time — the value is then taken from the environment variable. There is no environment-variable-only configuration channel and no env-only override.
 
 > **Kimi note**: Kimi has **two separate platforms with non-interchangeable API keys** — Moonshot (`https://api.moonshot.cn/v1`, keys `sk-...`, platform.moonshot.cn) and **Kimi For Coding** (`https://api.kimi.com/coding/v1`, keys `sk-kimi-...`, platform.kimi.com, model ID `k3`).
 > Use the `kimi` preset for Moonshot and `kimi-code` for Kimi For Coding — putting one platform's key on the other's endpoint fails with 401 (a hint is appended when the key/baseURL look mismatched).
@@ -148,6 +152,7 @@ Configuration comes exclusively from `~/.thincoder/config.json` — no environme
     "subagentModel": null, // default subagent provider/model override: "provider:model" | provider name | model name; null = inherit parent provider. Per-call: subagent tool `model` arg
     "subagentModels": {}, // per-type override: { "explore": "...", "plan": "...", "coder": "...", "eng-coder": "...", "eng-designer": "..." }; priority: tool model arg > this > subagentModel > parent provider
     "compactThreshold": 100000, // context compaction threshold (approx. tokens)
+    "timerWake": true, // timer idle wake: expired timers are delivered on the spot and start a turn when the host is idle — supported face = CLI / VSC / desktop foreground ends (headless chat / ACP: step-boundary delivery only); false = step-boundary delivery only
   },
   "memory": {
     // Path fields (dbPath / projectDir / team.dir / shell) expand a leading ~ (~, ~/, ~\) to the home directory at load time
@@ -196,7 +201,7 @@ src/
                          key handling, mouse/clipboard, the cmd-* command family, subagent panels
   tui.mjs                re-export shim → src/tui/index.mjs
   upgrade.mjs            version check and self-upgrade
-test/               node:test offline unit tests (npm test)
+test/               suite entry (npm test) — integration set; unit tests = batch-local files (docs/batches/) — current state: ../docs/core/design/TESTING.md §2.3
 scripts/            repo-side scripts — release check, doc-impact, syntax/endpoint probes,
                     real-environment verification (compaction, team sync)
 ```
@@ -213,7 +218,7 @@ Key design decisions:
 ## Development
 
 ```bash
-npm test                          # offline unit tests (node:test, with local mock servers)
+npm test                          # repo suite (integration set, offline) — empty manifest after the 2026-09-28 reset; unit tests = docs/batches/ — current state: ../docs/core/design/TESTING.md §2.3
 node scripts/verify-compress.mjs  # real-API verification of context compaction (needs valid config)
 node scripts/verify-team.mjs      # team memory A->git->B full-chain verification (local git, offline)
 ```
@@ -226,6 +231,17 @@ Code conventions: pure `.mjs`, no semicolons, no third-party npm dependencies al
 
 ## Changelog
 
+### 0.12.69 (2026-10)
+- **Provider-state unify (#841)** — one core resolution for the three ends (slot → `defaultModel` → first keyed channel) with three states: `ok`; `fallback` (runnable but no default model — one clear notice line, never blocked); `invalid` (no provider/key — guided picker; headless exits with a readable message and a clear code, while `fallback` prints one stderr line and keeps running). No more false "API key not configured" on a working channel.
+- **Core `^0.10.2`** — unified provider resolution (`resolveProviderPlan` / `resolveChannelModel`), `providerState` / `providerStateReason` keys, slot key gate.
+
+### 0.12.68 (2026-10)
+- **Idle wake** — suspended sessions wake to timers: they resume on schedule (visibility surface + abort accommodation).
+- **Ledger health signals** — an unhealthy ledger database now shows a warning line in the TUI (empty values render as `—`).
+- **Interaction batch (parity B1–B10)** — scroll exits, suspend queue, status line, render performance, i18n, settings, copy, stall indicator, send-while-busy queue.
+- **Core `^0.10.1`** — public-repo reading (declared sources), session-anchor resolution for multi-repo workspaces, the multi-repo operation mechanism, and discipline-surface fixes ride in with the core.
+- **Fixes** — tech-debt rounds 6/8 (slash passthrough, modal gate, sticky-bit re-arm, cascade clear, aborted reason, advisor slot guard, ledger writer); batch CRLF tolerance and paste lifecycle; ledger read/write guards.
+
 ### 0.12.67 (2026-09)
 - **Visible busy queue (multi-slot)** — queue up to 8 messages while a turn runs: a "to send" block shows each queued message in the stream, the status bar has four states, and batches are picked up at step boundaries (without interrupting in-flight tools) and merged into one message per turn (`/cmd` entries keep their order). New ledger commands: `thincoder ledger migrate --dry-run|--confirm` and `thincoder ledger audit`.
 - **Effort menus fixed** — the consult effort picker no longer shows a duplicated `none`; advisor effort "off" normalizes to `think_off` with an accurate receipt.

package/bin/thincoder.mjs

@@ -4,7 +4,7 @@
  * thincoder — CLI entry point
  *   thincoder                 Launch the interactive TUI
  *   thincoder chat "..."      One-shot agent run (tools enabled, streamed)
- *   thincoder memory <sub>    Memory management: list / search / put / remove
+ *   thincoder memory <sub>    Memory management: list / search / put / remove / sweep（sweep：origin 级库治理）
  *   thincoder upgrade         Update to the latest version from npm
  *   thincoder completion <sh> Shell completion: bash / zsh / fish
  *   thincoder session gc    Session dir GC: --dry-run report / --confirm delete cold projects (SESSION.md §6.12)
@@ -15,48 +15,56 @@
  */
 
 import { readFileSync } from "node:fs"
-import { isAbsolute, join } from "node:path"
-import { runAgent } from "@thincoder/core/agent.mjs"
 import { loadConfig, configPath } from "@thincoder/core/config.mjs"
 import { tracesRoot } from "@thincoder/core/traces/trace-store.mjs"
 import { scheduleTraceCleanup } from "@thincoder/core/traces/trace-cleanup.mjs"
-import { createMemory, syncDir } from "@thincoder/core/memory.mjs"
-// 批 1 CORE-DEFECT-FIXES B3：onWait 相位值域 + 文案单源（消费面禁各自枚举——PROVIDER.md §6.20）
-import { waitStatusText } from "@thincoder/core/provider/wait-status.mjs"
-import { assembleAgent, attachManifest, teamConfig, gitAuthor, validateProvider } from "../src/cli/make-agent.mjs"
-import { memoryCommand } from "../src/cli/memory-command.mjs"
-import { setupWizard } from "../src/cli/setup-wizard.mjs"
-import { summarize, askPermission } from "../src/cli/permission.mjs"
-import { distillCommand } from "../src/cli/distill-command.mjs"
-import { prepareCrashReporting, recentCrashHint, writeCrashRecord } from "../src/crash-reports.mjs"
-import { setTuiActive, restoreTerminalAfterCrash } from "../src/tui/tui-lifecycle.mjs"
+import { prepareCrashReporting, writeCrashRecord } from "../src/crash-reports.mjs"
+import { setTuiActive, restoreTerminalAfterCrash, _setCleanupOutPathForTest } from "../src/tui/tui-lifecycle.mjs"
 import { spawnTuiWrapped } from "../src/tui/wrapped-spawn.mjs"
 import { configurePromptInjections } from "@thincoder/core/prompt-files.mjs"
 import { CLI_PROMPT_INJECTIONS } from "../src/prompt-injections.mjs"
+// 拆档批 R4（2026-09-28）：命令分发表外提——分发骨架 + 八薄命令族 + help ∕ version =
+// `src/command-table.mjs`；交互长驻三命令（chat ∕ tui ∕ acp）= `src/command-interactive.mjs`。
+import { runCommandTable } from "../src/command-table.mjs"
 
 // U2（CORE-UNIFICATION §2.6.3 专项补⑦ / §2.13.8（六））：CLI 端锚取值表——**进程入口、任何装配之前**注册一次（调用期应用 ⇒ 无导入序要求）。
 // 漏配 = 锚字面静默进模型 ⇒ 入口面用例（test/integration/cli-prompt-entry.test.mjs）显式覆盖本调用路径。
 configurePromptInjections(CLI_PROMPT_INJECTIONS)
 
-const [command, ...args] = process.argv.slice(2)
+// CONFIG.md §6.2 ④：内部 argv 标志（生产零路径）。`--tui-wrapped`（包装门——注入点 =
+// src/tui/wrapped-spawn.mjs）在 bin 顶部自剥离（不进命令解析）；测试钩逐旗标
+// （`--test-crash` / `--test-tui-active` / `--test-cleanup-out=<路径>`）在下方崩溃钩处按原始 argv
+// （位置无关）判定后抛错接管——先于命令分派，永不作为命令被解释。
+const _argvRaw = process.argv.slice(2)
+const _tuiWrapped = _argvRaw.includes("--tui-wrapped")
+const [command, ...args] = _argvRaw.filter((a) => a !== "--tui-wrapped")
 // TUI-STDERR-CAPTURE（F-1）：TUI 启动（tui/无命令）默认包装——父 spawn 子 tee stderr 落盘（外部
 // 终止/native abort——第 4 类崩溃面——诊断默认捕获）。须在 prepareCrashReporting 前（父不预建/不设
-// report——子进程做——R25 保留）。env 门已设（包装内子进程）或包装失败 → 直行现逻辑（尽力面）。
-if ((command === undefined || command === "tui") && !process.env.THINCODER_TUI_WRAPPED) {
-  if (spawnTuiWrapped()) await new Promise(() => {}) // 包装成功 → 挂起（tee/收尾退出全在 wrapped-spawn——不达下方 switch）
+// report——子进程做——R25 保留）。argv 门已设（包装内子进程）或包装失败 → 直行现逻辑（尽力面）。
+if ((command === undefined || command === "tui") && !_tuiWrapped) {
+  if (spawnTuiWrapped()) await new Promise(() => {}) // 包装成功 → 挂起（tee/收尾退出全在 wrapped-spawn——不达下方命令分发）
 }
 const VERSION = JSON.parse(readFileSync(new URL("../package.json", import.meta.url), "utf8")).version
 
+// CONFIG.md §6.2 ①（用户面两开关——默认各向）：判定单点 = 本入口——读配置键
+// `diagnostics.{heapSnapshot,heapWatch}` 一次，经显式参数传入下方两消费者；读抛 ⇒ 各向默认
+// （watch 开 ∥ 快照关——fail-closed）。
+let _diagnostics = {}
+try {
+  _diagnostics = loadConfig().diagnostics ?? {}
+} catch { /* 配置缺失/损坏 → 各向默认（watch 开 ∥ 快照关——fail-closed） */ }
+
 // R25（F-R25b）：crash-reports 预建 + process.report 启用——入口最前（一切重活前——缩编程
 // 期窗口）——V8 OOM/原生 fatal 自动写 report.*.json（实现批实测：目录缺失时 Node 静默不写
 // ——预建为必要动作）。失败不阻断启动（尽力面）。
-prepareCrashReporting()
+prepareCrashReporting({ heapSnapshot: _diagnostics.heapSnapshot === true })
 // TUI-OOM-ROOTCAUSE（CRASH-REPORTS.md §8.3 武装点——全命令同源单点）：堆遥测/看门狗——
 // 60s 采样 + 双档（70/85%）比例边缘预警（stderr + TUI 行 + 事件日志）；定时器 unref
-// （一次性命令自然退出零阻塞）；默认开 / THINCODER_HEAP_WATCH 关值集可不启。失败面全吞。
+// （一次性命令自然退出零阻塞）；默认开（配置键 `diagnostics.heapWatch` 取 `false` 可不启）。
+// 失败面全吞。
 try {
   const { startHeapWatch } = await import("../src/heap-watch.mjs")
-  startHeapWatch()
+  startHeapWatch({ enabled: _diagnostics.heapWatch !== false })
 } catch { /* 看门狗启动失败不阻断启动（尽力面） */ }
 
 // R25（F-R25a）异常钩子升级：原"只 console.error 一行（TUI 全屏下不可见）"→ ① 落盘
@@ -91,12 +99,15 @@ function handleFatal(type, error) {
 process.on("uncaughtException", (error) => handleFatal("uncaughtException", error))
 process.on("unhandledRejection", (error) => handleFatal("unhandledRejection", error))
 
-// R25 测试门（T-R25a.1/a.2——子进程 env 注入——生产零路径）：THINCODER_TEST_CRASH=1 抛
-// 未捕获异常走完整崩溃序列；THINCODER_TEST_TUI_ACTIVE=1 模拟 TUI 活动态（同一 setter——
-// 测试缝同源）；THINCODER_TEST_CLEANUP_OUT 指向文件时恢复序列写入该文件（tui-lifecycle 读）。
-if (process.env.THINCODER_TEST_CRASH === "1") {
-  if (process.env.THINCODER_TEST_TUI_ACTIVE === "1") setTuiActive(true)
-  throw new Error("R25 test crash (THINCODER_TEST_CRASH)")
+// R25 测试门（T-R25a.1/a.2——子进程 argv 注入——生产零路径）：`--test-crash` 抛未捕获异常
+// 走完整崩溃序列；`--test-tui-active` 模拟 TUI 活动态（同一 setter——测试缝同源）；
+// `--test-cleanup-out=<路径>` 指到文件时恢复序列写入该文件（tui-lifecycle 缝——启动时读点）。
+// 判定面 = 原始 argv（位置无关——与旧 env 门同语义：旗标可居首/居尾，不依赖命令位）。
+const _cleanupOutArg = _argvRaw.find((a) => a.startsWith("--test-cleanup-out="))
+if (_cleanupOutArg) _setCleanupOutPathForTest(_cleanupOutArg.slice("--test-cleanup-out=".length))
+if (_argvRaw.includes("--test-crash")) {
+  if (_argvRaw.includes("--test-tui-active")) setTuiActive(true)
+  throw new Error("R25 test crash (--test-crash)")
 }
 
 const USAGE = `thincoder - thin coding agent
@@ -111,6 +122,7 @@ Usage:
   thincoder memory search <query>              Search memory
   thincoder memory put --type=<t> --title=<t> --content=<c> [--tags=<t>]
   thincoder memory remove <id>                 Remove an entry
+  thincoder memory sweep [--origin <o>] [--dry-run|--confirm]
   thincoder sync              Sync team memory repo (pull --rebase + reindex)
   thincoder reindex           Rebuild the local index from markdown sources
   thincoder distill <file> [--yes] [--layer=<s>]
@@ -161,321 +173,6 @@ if (command === undefined || command === "tui" || command === "chat" || command
   } catch { /* 拍挂点失败静默——不阻断启动 */ }
 }
 
-switch (command) {
-  case "chat": {
-    const auto = args.includes("--auto")
-    const prompt = args.filter((a) => a !== "--auto").join(" ").trim()
-    if (!prompt) {
-      console.error('Usage: thincoder chat [--auto] "<prompt>"')
-      exitSoon(1)
-      break
-    }
-
-    // R25（F-R25c）：非交互/chat 模式——上次异常终止提示写 stderr 一行（无匹配不提示）
-    const crashNotice = recentCrashHint()
-    if (crashNotice) console.error(crashNotice)
-
-    const agent = await assembleAgent()
-    // SESSION.md §6.8 D-S4（F4）+ MODEL-MERGE-SESSION：headless 无 TUI —— 可读错误 + 退出码 1，
-    // 不弹 UI、不崩溃；文案改引 defaultModel（F-6 引导 A headless 面——/config → 默认模型）
-    if (agent._providerInvalid) {
-      const dm = agent.config?.defaultModel
-      console.error(`[error] 未配置有效模型（defaultModel "${dm ?? "(not set)"}"：${agent._providerInvalidReason}）。请运行 thincoder 进入 TUI 设置（/config → 默认模型；或 /model 选择后以会话槽生效），或编辑 ${configPath}`)
-      exitSoon(1)
-      break
-    }
-    if (!agent.provider.apiKey) {
-      if (!process.stdin.isTTY) {
-        console.error(noKeyMessage())
-        exitSoon(1)
-        break
-      }
-      const p = await setupWizard()
-      if (!p) {
-        exitSoon(1)
-        break
-      }
-      agent.provider = p
-      agent.activeProvider = p.name
-      // Wizard may have configured an embedding key: attach vector search
-      const fresh = loadConfig()
-      if (fresh.embedding?.apiKey && agent.memory && !agent.memory.embedder) {
-        const { createEmbedder } = await import("@thincoder/core/embedding.mjs")
-        agent.memory.embedder = createEmbedder(fresh.embedding)
-      }
-    }
-    if (auto) agent.autoApprove = true
-    // Accumulate token usage, output to stderr at the end (don't pollute stdout pipe)
-    const usageTotal = { prompt: 0, completion: 0, cacheHit: 0, cacheMiss: 0 }
-    try {
-      await runAgent(agent, prompt, {
-        onToken: (text) => process.stdout.write(text),
-        onWait: (ev) => {
-          const s = waitStatusText(ev)
-          if (s) console.error(`[rate-limit] ${s}`)
-        },
-        onToolCall: (name, toolArgs) => {
-          console.error(`\n[tool] ${name} ${summarize(toolArgs)}`)
-        },
-        onToolResult: (name, result) => {
-          const preview = result.length > 200 ? result.slice(0, 200) + "..." : result
-          console.error(`[done] ${name} -> ${preview.split("\n")[0]}`)
-        },
-        onToolOutput: (name, chunk) => process.stderr.write(chunk),
-        onCompress: () => console.error(`\n[context] Context too long, auto-compacted (early conversation summarized by LLM)`),
-        onTaskUpdate: (items) => {
-          const done = items.filter((i) => i.status === "done").length
-          const current = items.find((i) => i.status === "in_progress")
-          console.error(`[task] ${done}/${items.length}${current ? ` ▶ ${current.title}` : ""}`)
-        },
-        onUsage: (usage) => {
-          usageTotal.prompt += usage.prompt_tokens ?? 0
-          usageTotal.completion += usage.completion_tokens ?? 0
-          usageTotal.cacheHit += usage.prompt_cache_hit_tokens ?? 0
-          usageTotal.cacheMiss += usage.prompt_cache_miss_tokens ?? 0
-        },
-        onPermissionRequest: (name, toolArgs) => (agent.autoApprove ? true : askPermission(name, toolArgs)),
-      })
-      process.stdout.write("\n")
-      if (usageTotal.prompt > 0) {
-        const cacheTotal = usageTotal.cacheHit + usageTotal.cacheMiss
-        const hitPart = cacheTotal > 0 ? ` cache-hit ${Math.round((usageTotal.cacheHit / cacheTotal) * 100)}%` : ""
-        console.error(`[usage] prompt ${usageTotal.prompt} + completion ${usageTotal.completion}${hitPart}`)
-      }
-    } catch (error) {
-      // 用 name 判断而非 instanceof：不依赖"与 runAgent 同一个模块实例"这一隐式约定
-      if (error.name === "ContinueError") {
-        console.error(`\n[paused] Agent stopped after ${error.turn} turns. Run in TUI to continue.`)
-        exitSoon(0)
-      } else {
-        console.error(`\n[error] ${error.message}`)
-        exitSoon(1)
-      }
-    }
-    break
-  }
-
-  case "memory": {
-    const config = loadConfig()
-    const memory = createMemory({ dbPath: config.memory.dbPath })
-    if (config.embedding?.apiKey) {
-      const { createEmbedder } = await import("@thincoder/core/embedding.mjs")
-      memory.embedder = createEmbedder(config.embedding)
-    }
-    if (config.memory.projectDir) {
-      memory.projectOrigin = isAbsolute(config.memory.projectDir) ? config.memory.projectDir : join(process.cwd(), config.memory.projectDir)
-    }
-    const code = await memoryCommand(memory, args)
-    if (code) exitSoon(code)
-    break
-  }
-
-  case "sync": {
-    const config = loadConfig()
-    const team = teamConfig(config)
-    if (!team) {
-      console.error("Team memory not configured. Set memory.team in ~/.thincoder/config.json:")
-      console.error('  "team": { "name": "myteam", "repo": "git@github.com:org/team-memory.git" }')
-      exitSoon(1)
-      break
-    }
-    const memory = createMemory({ dbPath: config.memory.dbPath })
-    const { ensureClone, pullTeam } = await import("@thincoder/core/git/gitmem.mjs")
-    try {
-      const cloned = await ensureClone(team)
-      if (cloned) console.log(`Cloned team repo to ${team.dir}`)
-      await pullTeam(team.dir)
-      const stats = await syncDir(memory, { layer: "team", dir: team.dir })
-      console.log(`Synced. Index: +${stats.added} ~${stats.updated} -${stats.removed}`)
-    } catch (error) {
-      console.error(`[error] ${error.message}`)
-      exitSoon(1)
-    }
-    break
-  }
-
-  case "distill": {
-    const code = await distillCommand(args, exitSoon)
-    if (code) exitSoon(code)
-    break
-  }
-
-  case "reindex": {
-    const config = loadConfig()
-    const memory = createMemory({ dbPath: config.memory.dbPath })
-    // files 层（project/team 的 markdown 索引）全量重建；索引是易失品，真相在 markdown
-    memory.db.prepare(`DELETE FROM files`).run()
-    const cwd = process.cwd()
-    let total = { added: 0, removed: 0 }
-    if (config.memory.projectDir) {
-      const s = await syncDir(memory, { layer: "project", dir: isAbsolute(config.memory.projectDir) ? config.memory.projectDir : join(cwd, config.memory.projectDir) })
-      total.added += s.added
-    }
-    const team = teamConfig(config)
-    if (team) {
-      const { ensureClone } = await import("@thincoder/core/git/gitmem.mjs")
-      await ensureClone(team)
-      const s = await syncDir(memory, { layer: "team", dir: team.dir })
-      total.added += s.added
-    }
-    console.log(`Reindexed ${total.added} markdown entries (project${team ? " + team" : ""}). Vectors will be lazily regenerated on next search.`)
-    break
-  }
-
-  case "tui":
-  case undefined: {
-    // 恢复上次的会话（同一项目目录）；provider 按保存的名字切回（用户上次可能换过模型）
-    // 2026-09-05 §6.10（R4）：恢复决策按本端记录 resumeSlot（D-2 ①②③）——manifest active
-    // 只作"无记录端"的一次性继承源，不再作本端恢复第一依据（D-6）。
-    // 2026-09-18（装配门禁批 #30 · KD-M1-13 / KD-M1-15）：resumeSlot **前移至装配之前**——
-    // 装配期模式门判据 = 会话权威值（槽优先 + config 回退）；恢复记录经 slotData 形参进
-    // 装配钩子（同进程同一调用前移 ⇒ 零新增写动作；钩子内不调 resumeSlot——非纯读）。
-    const { resumeSlot, applySession, sessionDescriptor } = await import("@thincoder/core/session.mjs")
-    // F-MI7：resumeSlot = async（探测束不阻塞事件循环）——此处必须 await
-    const { slot, data } = await resumeSlot(process.cwd())
-    const agent = await assembleAgent({ slotData: data })
-    // SESSION.md §6.8 D-S1：TUI 路径在 startTUI 前清空无效 provider——空 provider 不流入 runAgent
-    // （崩溃源：chat() 缺 model → 网关 400 或 fetch("undefined/...") TypeError）
-    if (agent._providerInvalid) agent.provider = null
-    const config = loadConfig()
-    if (data) {
-      // applySession 内部已按槽复合重算 compactThreshold（auto 时）——不再需要 switched 分支
-      // TUI-OOM-ROOTCAUSE（SESSION.md §6.14）：传 slot → 绑定记录存储（身份核验 + 对账）
-      // ——人读线 = 尾窗（磁盘为准）；未传 = 模式 F（全量数组）
-      applySession(agent, data, { slot })
-      // SESSION.md §6.11（2026-09-08——N6 评审 🔴 修复）：process restarted 句 = 进程级
-      // 信号——仅本启动 resume 路径设一次（真进程重启恢复盘上会话）；/session 切换与 ACP
-      // 加载走同一 applySession 收敛但不设（不误报进程重启）。prepareRun 发句即清——进程
-      // 内一次。resumed:yes 独立由 applySession 的恢复事件武装（data.history 非空）。
-      agent._processRestartPending = true
-    }
-    // D-3 钉槽：applySession 清 _slot 后立即钉回恢复槽——首保存必落恢复槽（消除
-    // "load → 首回合保存"窗口内并发方翻 active 导致的静默迁移）；/new 经 resetSessionState
-    // 清 _slot（newSession 已认领 + 写记录，语义保留）
-    agent._slot = slot
-    // M1 重估点（会话起点②——applySession 之后 · KD-M1-13 取值点②）：CLI 的工程模式权威值 =
-    // 会话槽（/eng 只写槽不写 config.json）；此时 agent.config.agent.engineering 已由
-    // applySession 按槽订正 ⇒ 与装配期判据同值 ⇒ 幂等重估（值同 ⇒ 无副作用）。
-    attachManifest(agent)
-    // D-S3 优先级补全：applySession 可能已用会话中的有效 provider 修复（config 无效 + 会话有效）——
-    // 修复后复验清除标记，仅当两者都无效才弹重选（validateProvider 幂等）
-    if (agent._providerInvalid) validateProvider(agent)
-    // MCP 连接失败在 TUI alt-buffer 下 stderr 不可见，注入为下一条 user 消息后的提醒
-    if (agent._mcpWarnings?.length) {
-      agent._pendingReminders = agent._pendingReminders ?? []
-      agent._pendingReminders.push(
-        `[System reminder: ${agent._mcpWarnings.length} MCP server(s) failed to connect at startup:\n` +
-        agent._mcpWarnings.map((w) => `  - ${w}`).join("\n") +
-        `\nYou can try reconnecting with /mcp connect <name>.]`
-      )
-    }
-    const { startTUI } = await import("../src/tui.mjs")
-    try {
-      await startTUI(agent, {
-        projectDir: config.memory.projectDir ? (isAbsolute(config.memory.projectDir) ? config.memory.projectDir : join(process.cwd(), config.memory.projectDir)) : null,
-        team: teamConfig(config),
-        author: gitAuthor(),
-        restored: data ? sessionDescriptor(agent, data) : data,
-        // R25（F-R25c）：TUI 启动显示一行"上次运行异常终止"提示（showStartup 渲染——无匹配不传）
-        crashNotice: recentCrashHint() ?? undefined,
-      })
-    } catch (error) {
-      console.error(`[error] ${error.message}`)
-      exitSoon(1)
-    }
-    break
-  }
-
-  case "completion": {
-    const shell = args[0]
-    if (!["bash", "zsh", "fish"].includes(shell)) {
-      console.error(`Usage: thincoder completion <bash|zsh|fish>`)
-      exitSoon(1)
-      break
-    }
-    // §11.2 交付行数债（2026-09-08——500 行硬限触碰执行）：直写段逐字迁出
-    // src/completions.mjs（脚本字节不变——printCompletion 发射）
-    const { printCompletion } = await import("../src/completions.mjs")
-    printCompletion(shell)
-    break
-  }
-
-  case "upgrade": {
-    const pkg = JSON.parse(readFileSync(new URL("../package.json", import.meta.url), "utf8"))
-    const local = pkg.version
-    const { checkForUpdate } = await import("../src/upgrade.mjs")
-    const result = await checkForUpdate(local)
-    if (!result) {
-      console.error("[upgrade] Unable to query npm registry — check your network connection and that npm is installed")
-      exitSoon(1)
-      break
-    }
-    if (!result.newer) {
-      console.log(`ThinCoder ${local} is already the latest.`)
-    } else {
-      console.log(`Upgrading: ${local} → ${result.latest}`)
-      const { execSync } = await import("node:child_process")
-      execSync("npm install -g thincoder@latest", { stdio: "inherit" })
-      console.log(`Upgraded to ${result.latest}`)
-    }
-    break
-  }
-
-  case "acp": {
-    const { runAcpServer, runAcpLogin } = await import("../src/acp.mjs")
-    // §11.2 改法 4：`args` 语义 = 追加到已配置的 agent 调用——`--login` = 终端认证流程，
-    // 不进 stdio 服务模式（不带该旗标时行为零变）；退出码语义 = 0 成功 / 非 0 失败。
-    if (args.includes("--login")) {
-      exitSoon(await runAcpLogin())
-      break
-    }
-    await runAcpServer()
-    break
-  }
-
-  case "session": {
-    // SESSION.md §6.12 / §6.17：会话目录 GC 显式面（冷 cwd 90 天面 + 三合取存量组全量面）——
-    // 双端同面（2026-09-21 端差注销：VS Code 端命令 `thincoder.sessionGc` 走同一核数据面）。
-    // SESSION.md §6.19 D-SE46（2026-09-22 会话索引批）：`session index` = 派生索引库显式面
-    // （--status 读数 / --rebuild 全量重扫）——同一 `session` 分发族（壳侧分发，实现在核）。
-    if (args[0] === "index") {
-      const { runSessionIndex } = await import("@thincoder/core/session-index-cmd.mjs")
-      process.exitCode = await runSessionIndex(args)
-      break
-    }
-    const { runSessionGc } = await import("@thincoder/core/session-gc.mjs")
-    process.exitCode = await runSessionGc(args)
-    break
-  }
-
-  case "ledger": {
-    // LEDGER.md §2.2（2026-09-25 批）：台账存量收正面——变体键合并迁移（migrate）/ 残档审计（audit）；
-    // 核内实现（ledger-migrate.mjs），壳侧只分发（先例 = `session gc` 双档：--dry-run 只读 / --confirm 执行）。
-    if (args[0] === "migrate" || args[0] === "audit") {
-      const m = await import("@thincoder/core/ledger-migrate.mjs")
-      process.exitCode = await (args[0] === "migrate" ? m.runLedgerMigrate(args) : m.runLedgerAudit(args))
-      break
-    }
-    console.error("Usage: thincoder ledger migrate --dry-run | --confirm [--from <key>] | thincoder ledger audit [--root <dir>]")
-    exitSoon(1)
-    break
-  }
-
-  case "--help":
-  case "-h": {
-    process.stdout.write(USAGE)
-    break
-  }
-
-  case "--version":
-  case "-v": {
-    console.log(VERSION)
-    break
-  }
-
-  default: {
-    console.error(`Unknown command: ${command}\n`)
-    process.stdout.write(USAGE)
-    exitSoon(1)
-  }
-}
+// 命令分发（2026-09-28 拆档批 R4：分发表外提 `src/command-table.mjs`）——依赖经 ctx 注入：
+// USAGE ∕ VERSION ∕ 两 helper = 本进程入口装配面（消费方：分发骨架与交互长驻三命令）。
+await runCommandTable(command, args, { usage: USAGE, version: VERSION, exitSoon, noKeyMessage })

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "thincoder",
-  "version": "0.12.67",
+  "version": "0.12.69",
   "description": "Thin coding agent - zero third-party dependencies, no build step, Node.js native. Sharp code, zero bloat.",
   "keywords": [
     "ai",
@@ -20,7 +20,7 @@
     "node": ">=24"
   },
   "dependencies": {
-    "@thincoder/core": "^0.9.5"
+    "@thincoder/core": "^0.10.2"
   },
   "files": [
     "bin/",

package/src/acp/bridge.mjs

@@ -308,9 +308,12 @@ export function buildAcpCallbacks({ sessionId, notify, request, log = () => {},
       // 2026-09-08 D1：单形态按行号改（path + line/startLine/endLine + new_string——无
       // old_string）同样走 IDE 缓冲通道（hasLineParams 判定）——否则回落本地写盘会与
       // IDE 缓冲脱敏；editSingle → computeEditEntry 行号语义自动继承。
-      if (base === "edit" && canReadFile && canWriteFile && (Array.isArray(args?.edits) || (path && typeof args?.new_string === "string" && (typeof args?.old_string === "string" || hasLineParams(args))))) {
+      // #327（TOOLS.md §6.17 裁定 3）：路由判据**归一**——`edits` 真值判（与核 execute 同判据）：
+      // 真值 ⇒ 批量分支（共享容器守卫、同一错误面）；假值 ⇒ 单形态面——「桥径单形态应用」分支消除。
+      const hasEdits = Boolean(args?.edits)
+      if (base === "edit" && canReadFile && canWriteFile && (hasEdits || (path && typeof args?.new_string === "string" && (typeof args?.old_string === "string" || hasLineParams(args))))) {
         try {
-          const text = Array.isArray(args?.edits) ? await editBatch(args) : await editSingle(path, args)
+          const text = hasEdits ? await editBatch(args) : await editSingle(path, args)
           return { handled: true, result: text }
         } catch (e) {
           return { handled: true, result: `Error: ${e.message}` }

package/src/acp/handlers-session.mjs

@@ -18,6 +18,9 @@ import { loadManifest, normalizeCwd, bindRecordStore, newSession, saveManifest,
 // core `session.mjs` 未 re-export 该名——直引 manifest 档，先例 = `thincoder-vscode/src/extension/session-io.mjs`）。
 import { staleClaims } from "@thincoder/core/session-slots-manifest.mjs"
 import { PLAN_ENGINEERING_REFUSED } from "@thincoder/core/agent-tools/plan.mjs"
+// #363①（off 形族单源——MODEL-SPECS.md §16.2）：thinking 开关的取形引核（本档零副本）。
+import { thinkOffShape } from "@thincoder/core/think-off.mjs"
+import { providerSpec } from "@thincoder/core/config.mjs"
 import { ACP_ERRORS } from "./transport.mjs"
 
 /** ENG-PLAN-EXCLUSION（FR31 ② · E7「命令面逐面钉定」）：工程真值 = 核单源同键
@@ -85,7 +88,15 @@ export function applyConfigOption(agent, configId, value) {
     }
     case "thinking": {
       if (typeof value !== "boolean" || !agent.provider) return false
-      agent.provider.thinking = value ? { type: "enabled" } : { type: "disabled" }
+      // off 取形 = 核单源（MODEL-SPECS.md §16.2——#363①）：effort 族 ⇒ `null`（载荷门据其补发
+      // `reasoning_effort:"none"` = 唯一有效 off 路径）；其余族 ⇒ `{ type:"disabled" }`。旧形
+      // 全族一形 `{type:"disabled"}` 在 effort 族被载荷层 falsy 跳过 ⇒ 关思考静默失效。
+      // on 侧同引核（thinkEnabledValue——自定义开值族 MiniMax = "adaptive"）。
+      const spec = providerSpec(agent.provider)
+      agent.provider.thinking = value ? { type: spec.thinkEnabledValue ?? "enabled" } : thinkOffShape(spec)
+      // effort 族：`null` 标记在载荷层由「无显式档」门控（provider/core.mjs:193 显式档优先会
+      // 跳过标记）⇒ off 清档位（与 CLI `/think off` 同法——cmd-think.mjs `applyThink` off 支）。
+      if (!value) delete agent.provider.reasoningEffort
       return true
     }
     case "mode": {