billion-context-pi 0.1.53 → 0.1.54-pr.236.56

package/README.md

@@ -22,6 +22,8 @@ The model decides <em>when</em> and <em>what</em> to compress — not a hard lim
 
 ---
 
+> **Host support:** this plugin is for **Pi**. It does **not** support **OMP (oh-my-pi)** — on an OMP host it refuses to run. OMP users: use [billion-context](https://github.com/ranxianglei/billion-context) instead (`bili omp`). Details: [docs/omp.md](./docs/omp.md).
+
 ## Why?
 
 When conversations get long, the model runs out of context. Most tools hard-truncate — silently dropping earlier messages. **billion-context** gives the model a `compress` tool: the LLM decides **when** and **what** to compress into high-fidelity summaries, preserving critical details (file paths, decisions, error strings) while reclaiming context space.
@@ -72,6 +74,22 @@ This has two practical implications:
 
 2. **Even with a single compression plugin, interference is still possible in rare cases.** Load order under Pi is determined by filesystem discovery order (`fs.readdirSync` over `.pi/extensions/` → global → packages), which is not fully deterministic. If another (non-compression) extension also hooks the `context` event and happens to load *after* billion-context-pi, it could modify the compressed output. billion-context-pi rebuilds its working set from the session log rather than the chained input, which makes it robust to handlers that run *before* it — but it cannot defend against a handler that runs *after* it. This is a limitation of Pi's extension model; if you observe unexpected context behavior, check whether other installed extensions intercept the `context` event.
 
+## Host support
+
+billion-context-pi is built for the **Pi** coding agent (`@earendil-works/pi-coding-agent`) and detects the host at session start:
+
+- **Pi** — fully supported.
+- **OMP (`can1357/oh-my-pi`)** — **not supported.** OMP's in-process session API diverges from Pi's, so the compression refs the extension injects can drift out of sync with the session's real refs and `compress` calls fail with `does not exist in this session` (issue [#234](https://github.com/ranxianglei/billion-context-pi/issues/234)). On OMP the extension now **refuses service**: it prints a warning, disables the ACP tools, and leaves the host's own context handling untouched.
+
+  **Use [billion-context](https://github.com/ranxianglei/billion-context) instead** — it runs the same compression pipeline server-side in a proxy, so the refs never diverge:
+
+  ```bash
+  npm install -g billion-context
+  bili omp   # run OMP through the proxy
+  ```
+
+  Full details: [docs/omp.md](./docs/omp.md).
+
 ## Model-facing tools
 
 | Tool | What it does |

package/README.zh-CN.md

@@ -20,6 +20,8 @@
 
 ---
 
+> **宿主支持:** 本插件面向 **Pi**。它**不支持 OMP(oh-my-pi)** —— 在 OMP 宿主上会拒绝运行。OMP 用户请改用 [billion-context](https://github.com/ranxianglei/billion-context)(`bili omp`)。详细说明:[docs/omp.zh-CN.md](./docs/omp.zh-CN.md)。
+
 ## 为什么选择 billion-context
 
 当对话变长,模型的上下文会耗尽。多数工具采用硬截断 —— 静默丢弃早期消息。**billion-context** 把 `compress` 工具交给模型:由 LLM 决定**何时**压缩、压缩**什么**,将内容压缩成高保真摘要,在回收上下文空间的同时保留关键细节(文件路径、决策、错误字符串)。
@@ -71,6 +73,22 @@ billion-context 通过拦截 Pi 的 `context` 事件接管上下文管理。**Pi
 
 2. **即使只有一个压缩插件,在少数情况下仍可能出现干扰。** Pi 下的加载顺序由文件系统发现顺序(`fs.readdirSync` 遍历 `.pi/extensions/` → 全局 → 包)决定,并不完全确定。如果另一个(非压缩类)扩展也 hook 了 `context` 事件、且恰好加载在 billion-context-pi *之后*,它可能修改压缩后的输出。billion-context-pi 从会话日志重建工作集(而非链式输入),这让它对*排在它之前*的 handler 鲁棒 —— 但无法防御*排在它之后*的 handler。这是 Pi 扩展模型的固有限制;若你观察到上下文行为异常,请检查是否有其他已安装扩展拦截了 `context` 事件。
 
+## 宿主支持
+
+billion-context-pi 面向 **Pi** 编码代理(`@earendil-works/pi-coding-agent`)构建,并在会话开始时检测宿主:
+
+- **Pi** — 完全支持。
+- **OMP(`can1357/oh-my-pi`)** — **不支持。** OMP 的进程内会话 API 与 Pi 不同,扩展注入的压缩引用可能与会话的真实引用漂移失步,导致 `compress` 调用失败,报错 `does not exist in this session`(issue [#234](https://github.com/ranxianglei/billion-context-pi/issues/234))。在 OMP 上,扩展现在会**拒绝服务**:打印警告、禁用 ACP 工具,并保持宿主自身的上下文处理不受影响。
+
+  **请改用 [billion-context](https://github.com/ranxianglei/billion-context)** — 它把同样的压缩流水线运行在服务端代理里,因此引用不会漂移:
+
+  ```bash
+  npm install -g billion-context
+  bili omp   # 让 OMP 通过代理运行
+  ```
+
+  完整说明:[docs/omp.zh-CN.md](./docs/omp.zh-CN.md)。
+
 ## 模型工具
 
 | 工具 | 作用 |

package/dist/compress-tool.d.ts

@@ -1,6 +1,12 @@
-import { Type } from "typebox";
+import { Type, type Static } from "typebox";
 import type { ToolDefinition } from "@earendil-works/pi-coding-agent";
 import type { AcpRuntime } from "./runtime.js";
+declare const RangeSpec: Type.TObject<{
+    startId: Type.TString;
+    endId: Type.TString;
+    summary: Type.TString;
+    topic: Type.TOptional<Type.TString>;
+}>;
 declare const CompressParams: Type.TObject<{
     topic: Type.TOptional<Type.TString>;
     content: Type.TUnion<[Type.TArray<Type.TObject<{
@@ -11,7 +17,11 @@ declare const CompressParams: Type.TObject<{
     }>>, Type.TString]>;
     summaryMaxChars: Type.TOptional<Type.TNumber>;
 }>;
+type CompressArgs = Static<typeof CompressParams>;
 export declare function makeCompressTool(runtime: AcpRuntime): ToolDefinition<typeof CompressParams>;
+type RangeEntry = Static<typeof RangeSpec>;
+export declare function normalizeRanges(args: CompressArgs): RangeEntry[] | string;
+export declare function tailRepair(s: string): string | undefined;
 /** Success = completed run that created >= 1 block (partial range errors
  *  still count: progress was made). A 0-block panel must NOT be success —
  *  it would reset the retry counter while the emergency nudge re-fires,

package/dist/config.d.ts

@@ -53,6 +53,12 @@ export interface CompressConfig extends CompressSettings {
  * (live model context window, protected tools, state persistence).
  */
 export interface AdapterConfig {
+    /** Master switch. Default: true. Set `enabled: false` in acp.json (or
+     *  programmatically) to turn the whole adapter off — no tools, no system
+     *  prompt, no context transform — for models too small to handle ACP, where
+     *  Pi's native context management should run instead. Checked at extension
+     *  load; requires a Pi restart to take effect. */
+    enabled?: boolean;
     /** When omitted, the adapter reads `ctx.model.contextWindow` live each turn.
      *  Set explicitly for tests/headless runs. */
     modelContextLimit?: number;

package/dist/delegate-tool.d.ts

@@ -2,13 +2,14 @@ import { type ChildProcess, type SpawnOptions } from "node:child_process";
 import { Type, type Static } from "typebox";
 import type { AgentToolResult, ExtensionAPI, ExtensionContext, ToolDefinition } from "@earendil-works/pi-coding-agent";
 import { type Usage } from "./delegate-events.js";
+export declare function delegateStdinText(resumeFrom: boolean, task: string | undefined): string;
 export declare function delegateSpawnOptions(cwd: string, env: NodeJS.ProcessEnv): SpawnOptions;
 /** Resolve the pi CLI entry for delegate child processes.
  *  argv[1] is only the pi CLI under a CLI host; embedded hosts (e.g. pi-web)
  *  run the SDK inside another node process, so probe instead. Non-pi hosts
  *  (omp) keep argv[1] untouched. */
 export declare function resolvePiCliEntry(argv1: string, env?: NodeJS.ProcessEnv, piHost?: boolean): string;
-type RunStatus = "running" | "completed" | "failed" | "cancelled";
+export type RunStatus = "running" | "completed" | "failed" | "cancelled";
 interface DelegateRun {
     runId: string;
     agent: string;
@@ -18,12 +19,18 @@ interface DelegateRun {
     finishedAt?: number;
     status: RunStatus;
     exitCode?: number | null;
+    /** Exit signal when the child died by signal (exit code null), e.g. "SIGTERM". */
+    exitSignal?: NodeJS.Signals;
     child?: ChildProcess;
     result?: {
         code: number | null;
         file: string;
         body: string;
     };
+    /** Live activity log path (async json-stream runs only). */
+    activityFile?: string;
+    /** runId of the run this run resumed from (resumeFrom). */
+    resumedFrom?: string;
     consumed?: boolean;
     /** True once the close handler injected the result as a system
      *  notification (sendUserMessage succeeded). Lets a later wait() avoid
@@ -88,7 +95,8 @@ export declare function makeEventApplier(opts: {
 export declare function resolveWaitTimeoutMs(raw: number | undefined): number;
 declare const DelegateParams: Type.TObject<{
     agent: Type.TString;
-    task: Type.TString;
+    task: Type.TOptional<Type.TString>;
+    resumeFrom: Type.TOptional<Type.TString>;
     cwd: Type.TOptional<Type.TString>;
     model: Type.TOptional<Type.TString>;
     async: Type.TOptional<Type.TBoolean>;
@@ -104,6 +112,13 @@ declare const WaitParams: Type.TObject<{
 }>;
 export declare function accumulateUsage(a: Usage | undefined, b: Usage): Usage;
 export declare function makeDelegateTool(pi: ExtensionAPI): ToolDefinition<typeof DelegateParams>;
+export declare function formatRunResult(run: DelegateRun): string;
+/** "exit 0" / "exit 1" / "exit SIGTERM" (signal shown when the child was
+ *  killed and has no exit code) / "exit ?" (unknown). */
+export declare function exitLabel(code: number | null, signal?: NodeJS.Signals | null): string;
+/** Shared note for cancelled runs: their files are retained, so point the
+ *  model at the partial output and offer a resume. */
+export declare function cancelledFileNote(runId: string, file: string): string;
 /** Runs that reached a terminal state but whose result never reached the
  *  model: no parked waiter, not consumed by a tool result, and never injected
  *  as a notification. The model was promised a notification ("do NOT keep
@@ -157,11 +172,21 @@ export declare function buildCancelResult(run: DelegateRun, content: string, mod
 };
 export declare function makeDelegateWaitTool(_pi: ExtensionAPI): ToolDefinition<typeof WaitParams>;
 export declare function makeDelegateCancelTool(_pi: ExtensionAPI): ToolDefinition<typeof CancelParams>;
-export declare function buildChildArgs(args: DelegateArgs, rolePrompt: string, ctx: ExtensionContext): Promise<{
+export declare function buildChildArgs(args: DelegateArgs, rolePrompt: string, ctx: ExtensionContext, runId: string): Promise<{
     cliArgs: string[];
     tmpDir: string;
     isAsync: boolean;
     useJsonStream: boolean;
+    sessionFile: string | null;
 }>;
-export declare function injectResult(pi: ExtensionAPI, agent: string, runId: string, task: string, code: number | null, file: string, timedOut?: string, usage?: Usage, mode?: "merged" | "separate", usageAlreadyReported?: boolean, body?: string): boolean;
+/** Watchdog/EOF finalize arrives with code === null (the child was killed or
+ *  never exited). If a result was delivered (non-empty reply or stderr), the
+ *  run counts as completed (0); otherwise it stays null = genuine failure. */
+export declare function effectiveExitCode(code: number | null, output: string, stderr: string): number | null;
+/** status (set by finalize from the effective exit code) is the authority for
+ *  the FAILED/completed decision; the raw code is diagnostic display only
+ *  ("exit ?"), so the notification can never disagree with run.status. */
+export declare function injectResult(pi: ExtensionAPI, agent: string, runId: string, task: string, status: RunStatus, code: number | null, file: string, timedOut?: string, usage?: Usage, mode?: "merged" | "separate", usageAlreadyReported?: boolean, body?: string, activityFile?: string, signal?: NodeJS.Signals | null): boolean;
+/** Tail of an activity log for failure diagnostics ("" when missing/empty). */
+export declare function readActivityTail(file: string, maxChars?: number): Promise<string>;
 export {};