billion-context-pi 0.1.61 → 0.1.62-pr.347.141

package/dist/config.d.ts

@@ -1,4 +1,5 @@
 import { type Config, type Prompts } from "acp-kernel";
+import type { CompressReasoningConfig } from "./reasoning-drop.js";
 import type { ThrottleRetryConfig } from "./throttle-retry.js";
 /** Per-role delegate defaults. Lets long-lived automation pin a cheaper or
  *  more capable model and a thinking level per delegate role, so the main
@@ -114,6 +115,10 @@ export interface CompressSettings {
      *  window and would suppress every nudge. Maps to kernel
      *  nudge.minPressureBenefitTokens. */
     minPressureBenefitTokens?: number;
+    /** [#336] Drop oversized reasoning (thinking) from historical `compress`
+     *  tool calls — see CompressReasoningConfig in src/reasoning-drop.ts.
+     *  Merged field-wise (drop, threshold) across the three levels. */
+    reasoning?: CompressReasoningConfig;
 }
 /** Per-provider compression overrides. Carries the same tuning fields as the
  *  global level, plus an optional per-model map keyed by model id. */

package/dist/index.js

@@ -3911,7 +3911,11 @@ function mergeCompress(global, provider, model) {
     maxContextLimit: model?.maxContextLimit ?? provider?.maxContextLimit ?? global?.maxContextLimit,
     emergencyThresholdPercent: model?.emergencyThresholdPercent ?? provider?.emergencyThresholdPercent ?? global?.emergencyThresholdPercent,
     nudgeGrowthTokens: model?.nudgeGrowthTokens ?? provider?.nudgeGrowthTokens ?? global?.nudgeGrowthTokens,
-    minPressureBenefitTokens: model?.minPressureBenefitTokens ?? provider?.minPressureBenefitTokens ?? global?.minPressureBenefitTokens
+    minPressureBenefitTokens: model?.minPressureBenefitTokens ?? provider?.minPressureBenefitTokens ?? global?.minPressureBenefitTokens,
+    reasoning: {
+      drop: model?.reasoning?.drop ?? provider?.reasoning?.drop ?? global?.reasoning?.drop,
+      threshold: model?.reasoning?.threshold ?? provider?.reasoning?.threshold ?? global?.reasoning?.threshold
+    }
   };
 }
 function resolveCompress(compress, provider, modelId) {
@@ -3953,6 +3957,63 @@ function parsePercent(v2) {
   return Number(s);
 }
 
+// src/reasoning-drop.ts
+var DEFAULT_COMPRESS_REASONING = { drop: true, threshold: 2048 };
+function resolveReasoningDrop(cfg) {
+  let threshold = DEFAULT_COMPRESS_REASONING.threshold;
+  if (cfg?.threshold !== void 0) {
+    const t = cfg.threshold;
+    if (typeof t === "number" && Number.isFinite(t) && t >= 0) {
+      threshold = Math.floor(t);
+    } else {
+      logWarn("config", { event: "compress-reasoning-invalid", field: "threshold", value: t, fallback: DEFAULT_COMPRESS_REASONING.threshold });
+    }
+  }
+  return { drop: cfg?.drop !== false, threshold };
+}
+function isThinking(part) {
+  const p = part;
+  return p?.type === "thinking" && typeof p.thinking === "string";
+}
+function hasCompressCall(content) {
+  if (!Array.isArray(content)) return false;
+  return content.some((p) => {
+    const b2 = p;
+    return b2?.type === "toolCall" && b2.name === "compress";
+  });
+}
+function reasoningLength(content) {
+  if (!Array.isArray(content)) return 0;
+  return content.reduce((n, p) => isThinking(p) ? n + p.thinking.length : n, 0);
+}
+function dropCompressReasoning(messages, cfg) {
+  const { drop, threshold } = resolveReasoningDrop(cfg);
+  if (!drop || messages.length === 0) return messages;
+  try {
+    let lastUser = -1;
+    for (let i = 0; i < messages.length; i++) {
+      if (messages[i].role === "user") lastUser = i;
+    }
+    if (lastUser < 0) return messages;
+    let changed = false;
+    const out = messages.slice();
+    for (let i = 0; i < lastUser; i++) {
+      const msg = messages[i];
+      if (msg.role !== "assistant" || !Array.isArray(msg.content)) continue;
+      if (!hasCompressCall(msg.content)) continue;
+      if (reasoningLength(msg.content) <= threshold) continue;
+      out[i] = {
+        ...msg,
+        content: msg.content.filter((p) => !isThinking(p))
+      };
+      changed = true;
+    }
+    return changed ? out : messages;
+  } catch {
+    return messages;
+  }
+}
+
 // src/tag-tokens.ts
 function formatTokens3(tokens) {
   if (tokens < 1e3) return String(tokens);
@@ -5175,6 +5236,10 @@ function createRuntime(adapter) {
     const m2 = ctx.model;
     return resolveConfig(adapterRef, liveContextLimit(ctx), m2?.provider, m2?.id);
   }
+  function reasoningDropFor(ctx) {
+    const m2 = ctx.model;
+    return resolveReasoningDrop(resolveCompress(adapterRef.compress, m2?.provider, m2?.id).reasoning);
+  }
   async function reloadConfig(cwd) {
     let user;
     try {
@@ -5253,7 +5318,7 @@ function createRuntime(adapter) {
   }, nudgeShownFor: (k) => nudgeShownTurns.has(k), nudgeShownTokensFor: (k) => nudgeShownTokens.get(k), clearNudgeTracking: () => {
     nudgeShownTurns.clear();
     nudgeShownTokens.clear();
-  }, clearNudgeTokenStamps: () => nudgeShownTokens.clear(), noteCompressOutcomes, compressRetryCappedFor, clearCompressRetryTracking, liveContextLimit, configFor, reloadConfig, stateFor, save, acquireLock, overflowFor, overflowDrop, noteDeadCompress, clearDeadCompress, throttleFor, throttleDrop, noteTokenScale, dropTokenScale };
+  }, clearNudgeTokenStamps: () => nudgeShownTokens.clear(), noteCompressOutcomes, compressRetryCappedFor, clearCompressRetryTracking, liveContextLimit, configFor, reasoningDropFor, reloadConfig, stateFor, save, acquireLock, overflowFor, overflowDrop, noteDeadCompress, clearDeadCompress, throttleFor, throttleDrop, noteTokenScale, dropTokenScale };
 }
 
 // node_modules/typebox/build/system/memory/memory.mjs
@@ -16749,7 +16814,7 @@ async function statusReport(runtime, ctx) {
   const sentTokens = estimateTokens(coreMessages, coveredIds, imageTokens) + systemPromptTokens;
   const viewSentTokens = adjustedTokenCount(runtime.core, coreMessages, state, config, sentTokens, imageTokens, systemPromptTokens);
   const turn = runtime.core.processTurn({ messages: coreMessages, state, config, tokenCount: anchorStale ? viewSentTokens : Math.max(viewSentTokens, realUsage?.tokens ?? 0) });
-  const versionStr = "0.1.61" ? `billion-context-pi@${"0.1.61"}` : void 0;
+  const versionStr = "0.1.62" ? `billion-context-pi@${"0.1.62"}` : void 0;
   let text = buildStatusPanel({
     version: versionStr,
     tokenCount: sessionTokens,
@@ -17283,7 +17348,7 @@ async function autoInstallLatest(latest, extDirOverride) {
       "--no-fund",
       "--no-save"
     ];
-    const prevVersion = (await readPackageJson(join11(extDir, "package.json")))?.version ?? "0.1.61";
+    const prevVersion = (await readPackageJson(join11(extDir, "package.json")))?.version ?? "0.1.62";
     const { code, stderr } = await runNpmImpl(installArgs(latest), { cwd: npmDir, timeout: 6e4 });
     if (code !== 0) {
       logWarn("update", {
@@ -17301,7 +17366,7 @@ async function autoInstallLatest(latest, extDirOverride) {
     }
     const verify = await verifyInstall(npmDir, latest);
     if (!verify.ok) {
-      const rollbackTo = SEMVER_RE.test(prevVersion) ? prevVersion : "0.1.61";
+      const rollbackTo = SEMVER_RE.test(prevVersion) ? prevVersion : "0.1.62";
       logWarn("update", { event: "auto-install-verify-failed", latest, reason: verify.reason, rollbackTo });
       const rb = await runNpmImpl(installArgs(rollbackTo), { cwd: npmDir, timeout: 6e4 });
       logInfo("update", { event: "rollback", from: latest, to: rollbackTo, ok: rb.code === 0 });
@@ -17376,7 +17441,7 @@ async function checkForUpdate(autoUpdate, notify) {
     }
     const latest = await fetchLatestVersion(tag);
     if (!latest) return;
-    const current = runtimeVersion ?? "0.1.61";
+    const current = runtimeVersion ?? "0.1.62";
     const hasUpdate = isVersionNewer(latest, current);
     debug.event("update-check", {
       current,
@@ -17554,7 +17619,7 @@ function wireSessionLifecycle(pi, runtime, standDownIfProxied) {
     setDelegatePolicy(DEFAULT_DELEGATE_POLICY);
     const sid = ctx.sessionManager.getSessionId();
     const modelInfo = ctx.model;
-    logInfo("session", { event: "start", sid, cwd: ctx.cwd, debug: runtime.adapter.debug ?? null, version: true ? "0.1.61" : null, model: modelInfo?.id ?? null, modelApi: modelInfo?.api ?? null, contextWindow: modelInfo?.contextWindow ?? null });
+    logInfo("session", { event: "start", sid, cwd: ctx.cwd, debug: runtime.adapter.debug ?? null, version: true ? "0.1.62" : null, model: modelInfo?.id ?? null, modelApi: modelInfo?.api ?? null, contextWindow: modelInfo?.contextWindow ?? null });
     try {
       await runtime.reloadConfig(ctx.cwd);
       const delegateCfg = resolveDelegate(runtime.adapter);
@@ -17701,7 +17766,13 @@ function wireContextTransform(pi, runtime, standDownIfProxied) {
         activeAfter: turn.state.blocks.filter((b2) => b2.active).length
       });
       const originalById = collectOriginals(entries);
-      const rebuilt = coreOutToAgentMessages(turn.messages, originalById);
+      let rebuilt = coreOutToAgentMessages(turn.messages, originalById);
+      const reasoningDrop = runtime.reasoningDropFor(ctx);
+      const droppedThinking = dropCompressReasoning(rebuilt, reasoningDrop);
+      if (droppedThinking !== rebuilt) {
+        debug.event("reasoning-drop", { sid, dropped: droppedThinking.length, drop: reasoningDrop.drop, threshold: reasoningDrop.threshold });
+      }
+      rebuilt = droppedThinking;
       const debugOn2 = debug.enabled;
       const turnKey = lastUserMessageId(entries) ?? sid;
       const compressOutcomes = collectCompressOutcomes(entries, turnStartIndex(entries));