billion-context-omp 0.2.5 → 0.2.6

package/README.md

@@ -65,11 +65,11 @@ Each message gets an invisible `<acp>` ref tag (`m00001`, `m00002`, ...) visible
 
 **The session stream is the single source of truth.** Compress calls live in the stream itself: every compress tool call's arguments (ranges + summaries) are re-applied deterministically on each LLM call, on restart, and on resume — no sidecar state file to drift out of sync. Position ids (`p1..pN`) and model-facing refs (`m00001..`) are re-derived from the stream every turn; prefix rewrites (retry, rewind, host compaction) are detected and safely re-folded, with fingerprint guards against replaying a call onto the wrong messages.
 
-omp's built-in `/compact` is intercepted and replaced by an ACP model-summarized compaction that also preserves prior compress-call summaries — nothing is lost to the gap between the summary and the kept entries.
+omp's built-in `/compact` is **the host's feature** — it runs natively (user-initiated, between turns). ACP does not intercept it. Compression itself is the model's decision via the compress tool. If you want ACP to be the *only* compression authority, set `"compaction": { "enabled": false }` in omp settings to disable the host's auto-compact (the 80% threshold trigger); manual `/compact` still works whenever you want it.
 
 ## Plugin compatibility
 
-**Keep exactly one context-compression plugin installed.** If two compression extensions both rewrite the message list, they clobber each other's work — compressed ranges can be re-expanded or corrupted. omp's own `/compact` is already intercepted automatically by billion-context-omp, but any *third-party* compression/compaction extension should be uninstalled.
+**Keep exactly one context-compression plugin installed.** If two compression extensions both rewrite the message list, they clobber each other's work — compressed ranges can be re-expanded or corrupted. Any *third-party* compression/compaction extension should be uninstalled.
 
 ## Model-facing tools
 
@@ -123,7 +123,6 @@ Create `~/.omp/acp-omp.json` (global) and/or `<project>/.omp/acp-omp.json` (proj
   "debug": false,
   "autoUpdate": true,
   "modelContextLimit": 200000,
-  "compressModel": "zhipuai:glm-5.2",
   "toolBashDefaultTimeout": 60,
   "toolOutputMaxBytes": 200000,
   "compress": {
@@ -145,10 +144,9 @@ Create `~/.omp/acp-omp.json` (global) and/or `<project>/.omp/acp-omp.json` (proj
 | Key | Default | Description |
 |-----|---------|-------------|
 | `debug` | `false` | Enable verbose **debug-level** events in the log. The always-on log (lifecycle events, errors, warnings) is written regardless; `debug` only adds extra diagnostics. Also enabled by env `ACP_DEBUG=1`. |
-| `transformMode` | `"context"` (default) or `"provider"` — where the compression surgery intercepts. `provider` transforms the provider wire payload (no feedback re-entry; experimental). |
+| `transformMode` | `"provider"` (default since v0.2.6) or `"context"` — where the compression surgery intercepts. `provider` transforms the provider wire payload (request-local, structurally immune to feedback re-entry). `context` is the legacy rewrite-in-place mode. |
 | `autoUpdate` | `true` | On session start (throttled to one check per 3 minutes), check npm for a newer version and auto-install it. Disable to avoid all startup network calls. |
 | `modelContextLimit` | *(auto)* | Override the context limit (in tokens). Defaults to the model's `contextWindow`. |
-| `compressModel` | *(session model)* | `provider:modelId` used for `/compact` model-summarized compaction (e.g. `"zhipuai:glm-5.2"`). Defaults to the current session model when omitted. |
 | `toolBashDefaultTimeout` | `60` | Seconds injected into the `bash` tool when the model omits `timeout`. Without this a forgotten timeout can hang for thousands of seconds. `0` restores unbounded behavior. |
 | `toolOutputMaxBytes` | `200000` | Hard byte cap on tool result text (applied via the `tool_result` hook). Stops runaway output that omp's own caps can't catch. When it fires the model is told where the full output lives; set lower (e.g. `8192`) for a tighter context budget, or `0` to disable. |
 | `compress.maxContextLimit` | `"75%"` | Context usage threshold that triggers **forced compression** nudges (bypasses growth-gate + cadence). Accepts a ratio (`0.75`) or percent string (`"75%"`). Lower = compress earlier / more aggressively. |

package/README.zh-CN.md

@@ -65,11 +65,11 @@ billion-context 拦截 omp 的 `context` 事件（每次 LLM 调用前触发）
 
 **会话流是唯一真相源。** compress 调用本身就存在于流中：每次 compress 工具调用的参数（区间 + 摘要）在每次 LLM 调用、重启、resume 时被确定性重放——没有会漂移失步的独立状态文件。位置 id（`p1..pN`）和面向模型的 ref（`m00001..`）每轮从流重新推导；前缀改写（retry、rewind、宿主 compaction）会被检测并安全重折叠，指纹守卫防止把调用重放到错误的消息上。
 
-omp 内置的 `/compact` 被拦截，替换为 ACP 模型摘要式 compaction，它同时保留之前的 compress 调用摘要——摘要与保留条目之间的空隙不会丢任何东西。
+omp 内置的 `/compact` **是宿主的功能**——原生运行（用户手动发起、轮间执行），ACP 不拦截。压缩本身由模型通过 compress 工具主动决策。若想让 ACP 成为*唯一*压缩主体，在 omp settings 里设 `"compaction": { "enabled": false }` 关掉宿主自动压缩（80% 阈值触发）；手动 `/compact` 随时可用。
 
 ## 插件兼容性
 
-**只安装一个上下文压缩插件。** 若两个压缩扩展都改写消息列表，它们会互相破坏对方的工作——被压缩的区间可能被重新展开或损坏。omp 自己的 `/compact` 已被 billion-context-omp 自动拦截，但任何*第三方*压缩/compaction 扩展都应卸载。
+**只安装一个上下文压缩插件。** 若两个压缩扩展都改写消息列表，它们会互相破坏对方的工作——被压缩的区间可能被重新展开或损坏。任何*第三方*压缩/compaction 扩展都应卸载。
 
 ## 面向模型的工具
 
@@ -123,7 +123,6 @@ billion-context-omp 开箱即用，无需配置。可选键写入 JSON 配置文
   "debug": false,
   "autoUpdate": true,
   "modelContextLimit": 200000,
-  "compressModel": "zhipuai:glm-5.2",
   "toolBashDefaultTimeout": 60,
   "toolOutputMaxBytes": 200000,
   "compress": {
@@ -147,7 +146,7 @@ billion-context-omp 开箱即用，无需配置。可选键写入 JSON 配置文
 | `debug` | `false` | 开启**debug 级**详细日志事件。常开日志（生命周期事件、错误、警告）无论如何都会写；`debug` 只增加诊断信息。也可用环境变量 `ACP_DEBUG=1` 开启。 |
 | `autoUpdate` | `true` | 会话启动时（节流为每 3 分钟最多一次）检查 npm 是否有新版本并自动安装。关闭可避免所有启动期网络请求。 |
 | `modelContextLimit` | *(自动)* | 覆盖上下文上限（token 数）。默认取模型的 `contextWindow`。 |
-| `compressModel` | *(会话模型)* | `/compact` 模型摘要式压缩使用的 `provider:modelId`（如 `"zhipuai:glm-5.2"`）。缺省用当前会话模型。 |
+| `transformMode` | `"provider"`（v0.2.6 起默认）或 `"context"` | 压缩手术拦截位置。`provider` 变换 provider wire 载荷（请求局部，结构性免疫回灌环路）。`context` 是旧版原地改写模式。 |
 | `toolBashDefaultTimeout` | `60` | 模型省略 `timeout` 时注入 `bash` 工具的秒数。没有它，一次忘记的 timeout 可能挂起数千秒。`0` 恢复无限制。 |
 | `toolOutputMaxBytes` | `200000` | 工具结果文本的硬字节上限（经 `tool_result` 钩子实施）。拦截 omp 自身上限管不住的失控输出。触发时模型会被告知完整输出在哪；调低（如 `8192`）可更省上下文，`0` 禁用。 |
 | `compress.maxContextLimit` | `"75%"` | 触发**强制压缩**提醒的上下文用量阈值（绕过增长门控与节拍）。接受比例（`0.75`）或百分比字符串（`"75%"`）。越低 = 越早/越激进压缩。 |

package/dist/config.d.ts

@@ -27,12 +27,15 @@ export interface CompressConfig {
  * (live model context window, protected tools, state persistence).
  */
 export interface AdapterConfig {
-    /** Where the compression surgery intercepts (issue #52). "context" (default)
-     *  rewrites the context event — battle-tested, but omp's recap/subagent
-     *  pipelines can re-feed our output as input (feedback-view loops).
-     *  "provider" leaves the agent array untouched and transforms the WIRE
-     *  payload at before_provider_request — request-local, no re-entry; unknown
-     *  provider formats pass through untransformed (fail-open). */
+    /** Where the compression surgery intercepts (issue #52). "provider"
+     *  (default since v0.2.6) leaves the agent array untouched and transforms
+     *  the WIRE payload at before_provider_request — request-local, no
+     *  re-entry, structurally immune to omp's feedback-view loops (the recap /
+     *  subagent re-feed pathology, issues #22/#52). Unknown provider formats
+     *  pass through untransformed (fail-open). "context" rewrites the context
+     *  event — battle-tested legacy mode, kept for compat.
+     *  Existing setups pin the mode explicitly via `transformMode` in
+     *  ~/.omp/acp-omp.json; unset now resolves to "provider". */
     transformMode?: "context" | "provider";
     /** When omitted, the adapter reads `ctx.model.contextWindow` live each turn.
      *  Set explicitly for tests/headless runs. */

package/dist/index.js

@@ -3663,7 +3663,7 @@ function createRuntime(adapter) {
       if (view.length === 0) return;
       let stream = view;
       let wire = false;
-      if ((adapterRef.transformMode ?? "context") === "provider" && ctx.model?.api === "openai-completions") {
+      if ((adapterRef.transformMode ?? "provider") === "provider" && ctx.model?.api === "openai-completions") {
         const base = getSystemPromptText(ctx);
         const acp = buildAcpSystemPrompt(promptsRef);
         stream = viewToWireStream(view, base.includes(acp) ? base : `${base}
@@ -4903,7 +4903,7 @@ async function statusReport(runtime, ctx) {
   const coveredIds = collectCoveredMessageIds(state);
   const sentTokens = estimateTokens(coreMessages, coveredIds) + systemPromptTokens;
   const turn = runtime.core.processTurn({ messages: coreMessages, state, config, tokenCount: sentTokens });
-  const versionStr = "0.2.5" ? `billion-context-omp@${"0.2.5"}` : void 0;
+  const versionStr = "0.2.6" ? `billion-context-omp@${"0.2.6"}` : void 0;
   return buildStatusPanel({
     version: versionStr,
     tokenCount: sessionTokens,
@@ -5193,7 +5193,7 @@ async function checkForUpdate(autoUpdate, notify) {
     const data = await res.json();
     const latest = data.version;
     if (!latest) return;
-    const current = runtimeVersion ?? "0.2.5";
+    const current = runtimeVersion ?? "0.2.6";
     const hasUpdate = isNewer(latest, current);
     debug.event("update-check", {
       current,
@@ -5469,9 +5469,9 @@ var index_default = createAcpExtension();
 function wireSessionLifecycle(pi, runtime) {
   pi.on("session_start", async (_event, ctx) => {
     const sid = ctx.sessionManager.getSessionId();
-    logInfo("session", { event: "start", sid, cwd: ctx.cwd, debug: runtime.adapter.debug ?? null, version: true ? "0.2.5" : null });
+    logInfo("session", { event: "start", sid, cwd: ctx.cwd, debug: runtime.adapter.debug ?? null, version: true ? "0.2.6" : null });
     const selfPath = import.meta.url;
-    const conflict = stampAndDetect(selfPath, true ? "0.2.5" : null);
+    const conflict = stampAndDetect(selfPath, true ? "0.2.6" : null);
     if (conflict) {
       logWarn("instance", { event: "dual-instance", self: selfPath, other: conflict.path, otherPid: conflict.pid, otherVersion: conflict.version });
       try {
@@ -5638,7 +5638,7 @@ ${rendered.text}${example}`);
 }
 function wireContextTransform(pi, runtime) {
   pi.on("context", async (event, ctx) => {
-    if ((runtime.adapter.transformMode ?? "context") === "provider") {
+    if ((runtime.adapter.transformMode ?? "provider") === "provider") {
       debug.event("context-observer-skip", { sid: ctx.sessionManager.getSessionId(), msgs: event.messages?.length ?? 0 });
       return void 0;
     }
@@ -5649,7 +5649,7 @@ function wireContextTransform(pi, runtime) {
 }
 function wireProviderTransform(pi, runtime) {
   pi.on("before_provider_request", async (event, ctx) => {
-    if ((runtime.adapter.transformMode ?? "context") !== "provider") return void 0;
+    if ((runtime.adapter.transformMode ?? "provider") !== "provider") return void 0;
     const payload = event.payload;
     if (payload === null || typeof payload !== "object" || !Array.isArray(payload.messages)) return void 0;
     const sid = ctx.sessionManager?.getSessionId?.() ?? "";