billion-context-pi 0.1.58 → 0.1.59-pr.301.129

package/README.md

@@ -103,6 +103,8 @@ billion-context-pi is built for the **Pi** coding agent (`@earendil-works/pi-cod
 
   Full details: [docs/omp.md](./docs/omp.md).
 
+- **Coexisting with the [billion-context](https://github.com/ranxianglei/billion-context) wire proxy** — running both on the same session double-compresses every request (wasted tokens, nested summaries, two ref coordinate systems). This is prevented automatically: launcher paths (`bili pi`, …) export `BILLION_CONTEXT_PROXY`, and models whose `baseUrl` routes through the proxy (`…/bili/https://upstream…`) are detected at session start — in both cases billion-context-pi stands down with a warning and leaves the proxy as the sole compressor. One exception: transparent mode, where traffic reaches the proxy via `HTTPS_PROXY` so the URL carries no `/bili/` prefix — that is undetectable from the URL, so export `BILLION_CONTEXT_PROXY=1` before starting pi in that case.
+
 ## Model-facing tools
 
 | Tool | What it does |
@@ -222,10 +224,10 @@ The `.acp.json` sidecar is what holds your compressed blocks. Without it, the se
 
 ### Migrating a session (cross-machine copy / backup-restore)
 
-Pi's built-in export/import moves only the **transcript**, not the ACP state. Two things are lost:
+Pi's built-in export/import moves only the **transcript**, not the ACP state. Two things are affected:
 
-1. **The `.acp.json` sidecar is not carried.** An imported session therefore has *no* compressed blocks: every LLM call resends the entire raw history until the nudge re-compresses it — a one-time full re-cache cost plus context bloat back to original size. Very long sessions can approach or exceed the model window before re-compression kicks in (Pi's native compaction is disabled while this plugin is active, so ACP is the only context manager).
-2. **Export drops the `parentSession` header.** Clone/fork child sessions rely on that header field to inherit their parent's compression state; once exported, the link is gone even if the parent's files still exist on the target machine.
+1. **The `.acp.json` sidecar is not carried.** As a fallback, ACP now **rebuilds the compression state by replaying the session log itself**: on the first context event after import, it re-applies every successful `compress` call recorded in the transcript (assistant tool-call arguments + tool results) through the kernel, restoring the block structure, summaries, message refs and cumulative stats, then persists the recreated sidecar (#299). Errored, no-op and unparseable calls are skipped; the replay runs only when no sidecar and no inheritable parent state exists. This removes the old behavior — resending the entire raw history until the nudge re-compresses, with a one-time full re-cache cost and context bloat back to original size. Caveat: the replay restores what the *log* records, so summaries come back exactly as stored in the transcript, and stats like per-message token snapshots are re-derived rather than bit-identical.
+2. **Export drops the `parentSession` header.** Clone/fork child sessions rely on that header field to inherit their parent's compression state; once exported, the link is gone even if the parent's files still exist on the target machine. (A host-side fix is proposed upstream in [pi#1](https://github.com/ranxianglei/pi/pull/1).)
 
 **To migrate a session with its compression state intact, copy both files together** (they share the same base name):
 
@@ -238,7 +240,7 @@ cp -r ~/.pi/agent/sessions  <backup>/pi-sessions
 
 Restore them next to each other on the target machine. For clone/fork children, also bring the parent's pair so `parentSession` resolves.
 
-> A permanent fix — the host carrying the sidecar through import/export and preserving `parentSession` — belongs upstream in pi-coding-agent and is tracked in issue [#299](https://github.com/ranxianglei/billion-context-pi/issues/299). Until it lands, copy the pair manually.
+> If you import only the `.jsonl`, ACP's log-replay fallback rebuilds the state automatically on the next session (see above). Copying the pair is still preferred — it is exact, while the replay re-derives token snapshots and can only restore what the transcript records.
 
 ## Built on acp-kernel
 

package/README.zh-CN.md

@@ -102,6 +102,8 @@ billion-context-pi 面向 **Pi** 编码代理(`@earendil-works/pi-coding-agent`)
 
   完整说明:[docs/omp.zh-CN.md](./docs/omp.zh-CN.md)。
 
+- **与 [billion-context](https://github.com/ranxianglei/billion-context) 线代理共存** —— 两者同时作用于同一会话会对每个请求双重压缩(token 浪费、嵌套摘要、两套 ref 坐标系)。这会被自动防止:launcher 路径(`bili pi` 等)导出 `BILLION_CONTEXT_PROXY`;模型 `baseUrl` 经代理路由(`…/bili/https://upstream…`)时在会话开始即被检测到 —— 两种情况下 billion-context-pi 都会带警告让位,由代理独占压缩。一个例外:透明模式(流量经 `HTTPS_PROXY` 到达代理、URL 无 `/bili/` 前缀)无法从 URL 识别 —— 此时请在启动 pi 前导出 `BILLION_CONTEXT_PROXY=1`。
+
 ## 模型工具
 
 | 工具 | 作用 |

package/dist/config.d.ts

@@ -206,6 +206,12 @@ export interface AdapterConfig {
      *  disables) or a ThrottleRetryConfig object. Default: enabled, 10 retries,
      *  60s exponential base capped at 300s per kick. */
     throttleRetry?: boolean | ThrottleRetryConfig;
+    /** Cap on the output-headroom reservation as a fraction of the context
+     *  window: reserved = min(model.maxTokens, pct * window). Accepts a ratio
+     *  (0.25) or percent string ("25%"). Default: 0.25. Set 0 to disable the
+     *  reservation entirely; >= 1 restores the legacy full-capability
+     *  reservation (issue #207). */
+    outputHeadroomMaxPct?: number | string;
     /** Generic tool-call repetition guard (see RepetitionGuardConfig). Accepts a
      *  boolean shorthand (`false` disables) or an object. Default: enabled,
      *  warn=3, abort=5. Stops greedy small models looping on byte-identical