vivekmind 1.0.0 → 1.0.1

package/README.md

@@ -143,7 +143,8 @@ Over 40 built-in commands for session control, configuration, and workflow manag
 
 ```
 /model          Switch AI model
-/compress       Compress chat history to save context
+/context         Show context window usage breakdown (add "detail" for per-item)
+/compress        Compress chat history to save context (shows % saved)
 /clear          Clear conversation
 /memory         Open memory manager
 /remember       Save info to memory
@@ -260,6 +261,7 @@ vivekmind -p "refactor module" --output stream-json
 
 ### Additional Features
 
+- **Context Compression** — Aggressive multi-layer compression system that reduces token usage by 80-95% on long sessions, minimizing AWS Bedrock costs. Combines LLM-based summarization, microcompaction (automatic old tool output clearing), and output pruning to keep 100k+ token sessions within budget. Triggered automatically at 70% context usage or manually via `/compress`.
 - **Sandbox** — Docker and Podman sandbox support for safe code execution
 - **LSP Integration** — Language Server Protocol for code intelligence
 - **Extensions** — Install community extensions from GitHub or npm
@@ -275,6 +277,33 @@ vivekmind -p "refactor module" --output stream-json
 
 ---
 
+## Token Optimization
+
+VivekMind includes a comprehensive token optimization system designed to minimize API costs, especially important for AWS Bedrock users who pay per token.
+
+### How It Works
+
+The compression system operates on three layers:
+
+1. **Microcompaction (automatic)** — Before every API turn, old tool results (file reads, shell output, grep results) that are no longer relevant are replaced with a one-line placeholder. This is triggered by either idle time (>60 min) or when tool outputs consume more than 20% of the context window. The most recent 3 results are preserved.
+
+2. **Output Pruning (automatic)** — When building the compression prompt, high-volume tool outputs (file reads, shell, grep, glob, etc.) are aggressively truncated to 80 characters per tool in the history being summarized. This prevents the compression LLM call itself from bloating.
+
+3. **LLM-Based Compression (automatic + manual)** — When context usage exceeds 70% of the model's context window, the entire conversation history (except the most recent 5%) is sent to the LLM to produce a structured XML `<state_snapshot>` containing: overall goal, key knowledge, user preferences, file system state, unresolved bugs, recent actions, and current plan. The summary is capped at 1024 output tokens. Manual trigger via `/compress` uses a 3% preserve threshold for more aggressive compression.
+
+### Context Usage Display
+
+The footer shows real-time context window usage as a percentage with the actual token count (e.g., `45.2% context used (181.2k/200k tok)`). The `/context` command shows a detailed breakdown by category: system prompt, built-in tools, MCP tools, memory files, skills, and messages.
+
+### Cost Impact
+
+For a typical 100k-token session on AWS Bedrock (Claude Sonnet):
+- **Without compression**: Every turn sends the full 100k+ tokens, costing ~$1.50/turn
+- **With compression**: History is summarized to ~3-5k tokens after the 70% threshold, reducing per-turn cost by 80-95% for subsequent turns
+- **Microcompaction alone**: Clears 40-60% of historical tool output bloat before it reaches the LLM
+
+---
+
 ## Configuration
 
 ### Settings

package/cli.js

@@ -172983,7 +172983,7 @@ __export(geminiContentGenerator_exports, {
   createGeminiContentGenerator: () => createGeminiContentGenerator
 });
 function createGeminiContentGenerator(config2, gcConfig) {
-  const version2 = "1.0.0";
+  const version2 = "1.0.1";
   const userAgent2 = config2.userAgent || `VivekMindCode/${version2} (${process.platform}; ${process.arch})`;
   const baseHeaders = {
     "User-Agent": userAgent2
@@ -203095,7 +203095,7 @@ var init_pricing = __esm({
 });
 
 // packages/core/src/core/bedrockContentGenerator/bedrockContentGenerator.js
-var import_client_bedrock_runtime, debugLogger21, BedrockContentGenerator;
+var import_client_bedrock_runtime, DUMMY_TOOL, debugLogger21, BedrockContentGenerator;
 var init_bedrockContentGenerator = __esm({
   "packages/core/src/core/bedrockContentGenerator/bedrockContentGenerator.js"() {
     "use strict";
@@ -203107,6 +203107,18 @@ var init_bedrockContentGenerator = __esm({
     init_safeJsonParse();
     init_debugLogger();
     init_tokenLimits();
+    DUMMY_TOOL = {
+      toolSpec: {
+        name: "dummy_tool_prevent_validation_error",
+        description: "A placeholder tool to satisfy AWS Bedrock requirements when conversation history contains tool blocks.",
+        inputSchema: {
+          json: {
+            type: "object",
+            properties: {}
+          }
+        }
+      }
+    };
     debugLogger21 = createDebugLogger("BEDROCK");
     BedrockContentGenerator = class {
       static {
@@ -203139,12 +203151,15 @@ var init_bedrockContentGenerator = __esm({
         const modelId = this.contentGeneratorConfig.model;
         const { messages, system } = this.converter.convertGeminiRequestToConverse(request4);
         const toolConfig = request4.config?.tools ? await this.converter.convertGeminiToolsToConverse(request4.config.tools) : void 0;
+        const hasToolUseOrToolResult = messages.some(
+          (msg) => msg.content?.some((block2) => "toolUse" in block2 || "toolResult" in block2)
+        );
         const inferenceConfig = this.buildInferenceConfig(request4);
         const command2 = new import_client_bedrock_runtime.ConverseCommand({
           modelId,
           messages,
           system,
-          toolConfig: toolConfig && toolConfig.tools && toolConfig.tools.length > 0 ? toolConfig : void 0,
+          toolConfig: toolConfig && toolConfig.tools && toolConfig.tools.length > 0 ? toolConfig : hasToolUseOrToolResult ? { tools: [DUMMY_TOOL] } : void 0,
           inferenceConfig
         });
         debugLogger21.info(`Converse request: model=${modelId}, messages=${messages.length}`);
@@ -203162,12 +203177,15 @@ var init_bedrockContentGenerator = __esm({
         const modelId = this.contentGeneratorConfig.model;
         const { messages, system } = this.converter.convertGeminiRequestToConverse(request4);
         const toolConfig = request4.config?.tools ? await this.converter.convertGeminiToolsToConverse(request4.config.tools) : void 0;
+        const hasToolUseOrToolResult = messages.some(
+          (msg) => msg.content?.some((block2) => "toolUse" in block2 || "toolResult" in block2)
+        );
         const inferenceConfig = this.buildInferenceConfig(request4);
         const command2 = new import_client_bedrock_runtime.ConverseStreamCommand({
           modelId,
           messages,
           system,
-          toolConfig: toolConfig && toolConfig.tools && toolConfig.tools.length > 0 ? toolConfig : void 0,
+          toolConfig: toolConfig && toolConfig.tools && toolConfig.tools.length > 0 ? toolConfig : hasToolUseOrToolResult ? { tools: [DUMMY_TOOL] } : void 0,
           inferenceConfig
         });
         debugLogger21.info(`ConverseStream request: model=${modelId}, messages=${messages.length}`);
@@ -207083,7 +207101,7 @@ This error was probably caused by cyclic schema references in one of the followi
           }
           if (chunk.usageMetadata) {
             usageMetadata = chunk.usageMetadata;
-            const lastPromptTokenCount = usageMetadata.totalTokenCount || usageMetadata.promptTokenCount;
+            const lastPromptTokenCount = usageMetadata.promptTokenCount || usageMetadata.totalTokenCount;
             if (lastPromptTokenCount && this.telemetryService) {
               this.telemetryService.setLastPromptTokenCount(lastPromptTokenCount);
             }
@@ -367994,7 +368012,7 @@ var require_backend = __commonJS({
                 ) : symbolOrNumber;
               }
               __name(getTypeSymbol, "getTypeSymbol");
-              var _ReactTypeOfWork = ReactTypeOfWork, CacheComponent = _ReactTypeOfWork.CacheComponent, ClassComponent = _ReactTypeOfWork.ClassComponent, IncompleteClassComponent = _ReactTypeOfWork.IncompleteClassComponent, FunctionComponent = _ReactTypeOfWork.FunctionComponent, IndeterminateComponent = _ReactTypeOfWork.IndeterminateComponent, ForwardRef = _ReactTypeOfWork.ForwardRef, HostRoot = _ReactTypeOfWork.HostRoot, HostHoistable = _ReactTypeOfWork.HostHoistable, HostSingleton = _ReactTypeOfWork.HostSingleton, HostComponent = _ReactTypeOfWork.HostComponent, HostPortal = _ReactTypeOfWork.HostPortal, HostText = _ReactTypeOfWork.HostText, Fragment29 = _ReactTypeOfWork.Fragment, LazyComponent = _ReactTypeOfWork.LazyComponent, LegacyHiddenComponent = _ReactTypeOfWork.LegacyHiddenComponent, MemoComponent = _ReactTypeOfWork.MemoComponent, OffscreenComponent = _ReactTypeOfWork.OffscreenComponent, Profiler = _ReactTypeOfWork.Profiler, ScopeComponent = _ReactTypeOfWork.ScopeComponent, SimpleMemoComponent = _ReactTypeOfWork.SimpleMemoComponent, SuspenseComponent = _ReactTypeOfWork.SuspenseComponent, SuspenseListComponent = _ReactTypeOfWork.SuspenseListComponent, TracingMarkerComponent = _ReactTypeOfWork.TracingMarkerComponent;
+              var _ReactTypeOfWork = ReactTypeOfWork, CacheComponent = _ReactTypeOfWork.CacheComponent, ClassComponent = _ReactTypeOfWork.ClassComponent, IncompleteClassComponent = _ReactTypeOfWork.IncompleteClassComponent, FunctionComponent = _ReactTypeOfWork.FunctionComponent, IndeterminateComponent = _ReactTypeOfWork.IndeterminateComponent, ForwardRef = _ReactTypeOfWork.ForwardRef, HostRoot = _ReactTypeOfWork.HostRoot, HostHoistable = _ReactTypeOfWork.HostHoistable, HostSingleton = _ReactTypeOfWork.HostSingleton, HostComponent = _ReactTypeOfWork.HostComponent, HostPortal = _ReactTypeOfWork.HostPortal, HostText = _ReactTypeOfWork.HostText, Fragment28 = _ReactTypeOfWork.Fragment, LazyComponent = _ReactTypeOfWork.LazyComponent, LegacyHiddenComponent = _ReactTypeOfWork.LegacyHiddenComponent, MemoComponent = _ReactTypeOfWork.MemoComponent, OffscreenComponent = _ReactTypeOfWork.OffscreenComponent, Profiler = _ReactTypeOfWork.Profiler, ScopeComponent = _ReactTypeOfWork.ScopeComponent, SimpleMemoComponent = _ReactTypeOfWork.SimpleMemoComponent, SuspenseComponent = _ReactTypeOfWork.SuspenseComponent, SuspenseListComponent = _ReactTypeOfWork.SuspenseListComponent, TracingMarkerComponent = _ReactTypeOfWork.TracingMarkerComponent;
               function resolveFiberType(type) {
                 var typeSymbol = getTypeSymbol(type);
                 switch (typeSymbol) {
@@ -368040,7 +368058,7 @@ var require_backend = __commonJS({
                   case HostPortal:
                   case HostText:
                     return null;
-                  case Fragment29:
+                  case Fragment28:
                     return "Fragment";
                   case LazyComponent:
                     return "Lazy";
@@ -368106,7 +368124,7 @@ var require_backend = __commonJS({
             function attach(hook2, rendererID, renderer2, global2) {
               var version2 = renderer2.reconcilerVersion || renderer2.version;
               var _getInternalReactCons = getInternalReactConstants(version2), getDisplayNameForFiber = _getInternalReactCons.getDisplayNameForFiber, getTypeSymbol = _getInternalReactCons.getTypeSymbol, ReactPriorityLevels = _getInternalReactCons.ReactPriorityLevels, ReactTypeOfWork = _getInternalReactCons.ReactTypeOfWork, StrictModeBits = _getInternalReactCons.StrictModeBits;
-              var CacheComponent = ReactTypeOfWork.CacheComponent, ClassComponent = ReactTypeOfWork.ClassComponent, ContextConsumer = ReactTypeOfWork.ContextConsumer, DehydratedSuspenseComponent = ReactTypeOfWork.DehydratedSuspenseComponent, ForwardRef = ReactTypeOfWork.ForwardRef, Fragment29 = ReactTypeOfWork.Fragment, FunctionComponent = ReactTypeOfWork.FunctionComponent, HostRoot = ReactTypeOfWork.HostRoot, HostHoistable = ReactTypeOfWork.HostHoistable, HostSingleton = ReactTypeOfWork.HostSingleton, HostPortal = ReactTypeOfWork.HostPortal, HostComponent = ReactTypeOfWork.HostComponent, HostText = ReactTypeOfWork.HostText, IncompleteClassComponent = ReactTypeOfWork.IncompleteClassComponent, IndeterminateComponent = ReactTypeOfWork.IndeterminateComponent, LegacyHiddenComponent = ReactTypeOfWork.LegacyHiddenComponent, MemoComponent = ReactTypeOfWork.MemoComponent, OffscreenComponent = ReactTypeOfWork.OffscreenComponent, SimpleMemoComponent = ReactTypeOfWork.SimpleMemoComponent, SuspenseComponent = ReactTypeOfWork.SuspenseComponent, SuspenseListComponent = ReactTypeOfWork.SuspenseListComponent, TracingMarkerComponent = ReactTypeOfWork.TracingMarkerComponent;
+              var CacheComponent = ReactTypeOfWork.CacheComponent, ClassComponent = ReactTypeOfWork.ClassComponent, ContextConsumer = ReactTypeOfWork.ContextConsumer, DehydratedSuspenseComponent = ReactTypeOfWork.DehydratedSuspenseComponent, ForwardRef = ReactTypeOfWork.ForwardRef, Fragment28 = ReactTypeOfWork.Fragment, FunctionComponent = ReactTypeOfWork.FunctionComponent, HostRoot = ReactTypeOfWork.HostRoot, HostHoistable = ReactTypeOfWork.HostHoistable, HostSingleton = ReactTypeOfWork.HostSingleton, HostPortal = ReactTypeOfWork.HostPortal, HostComponent = ReactTypeOfWork.HostComponent, HostText = ReactTypeOfWork.HostText, IncompleteClassComponent = ReactTypeOfWork.IncompleteClassComponent, IndeterminateComponent = ReactTypeOfWork.IndeterminateComponent, LegacyHiddenComponent = ReactTypeOfWork.LegacyHiddenComponent, MemoComponent = ReactTypeOfWork.MemoComponent, OffscreenComponent = ReactTypeOfWork.OffscreenComponent, SimpleMemoComponent = ReactTypeOfWork.SimpleMemoComponent, SuspenseComponent = ReactTypeOfWork.SuspenseComponent, SuspenseListComponent = ReactTypeOfWork.SuspenseListComponent, TracingMarkerComponent = ReactTypeOfWork.TracingMarkerComponent;
               var ImmediatePriority = ReactPriorityLevels.ImmediatePriority, UserBlockingPriority = ReactPriorityLevels.UserBlockingPriority, NormalPriority = ReactPriorityLevels.NormalPriority, LowPriority = ReactPriorityLevels.LowPriority, IdlePriority = ReactPriorityLevels.IdlePriority, NoPriority = ReactPriorityLevels.NoPriority;
               var getLaneLabelMap = renderer2.getLaneLabelMap, injectProfilingHooks = renderer2.injectProfilingHooks, overrideHookState = renderer2.overrideHookState, overrideHookStateDeletePath = renderer2.overrideHookStateDeletePath, overrideHookStateRenamePath = renderer2.overrideHookStateRenamePath, overrideProps = renderer2.overrideProps, overridePropsDeletePath = renderer2.overridePropsDeletePath, overridePropsRenamePath = renderer2.overridePropsRenamePath, scheduleRefresh = renderer2.scheduleRefresh, setErrorHandler = renderer2.setErrorHandler, setSuspenseHandler = renderer2.setSuspenseHandler, scheduleUpdate = renderer2.scheduleUpdate;
               var supportsTogglingError = typeof setErrorHandler === "function" && typeof scheduleUpdate === "function";
@@ -368321,7 +368339,7 @@ var require_backend = __commonJS({
                     return true;
                   case HostRoot:
                     return false;
-                  case Fragment29:
+                  case Fragment28:
                     return key === null;
                   default:
                     var typeSymbol = getTypeSymbol(type);
@@ -368396,7 +368414,7 @@ var require_backend = __commonJS({
                     return ElementTypeHostComponent;
                   case HostPortal:
                   case HostText:
-                  case Fragment29:
+                  case Fragment28:
                     return ElementTypeOtherOrUnknown;
                   case MemoComponent:
                   case SimpleMemoComponent:
@@ -467989,7 +468007,7 @@ var require_react_is_development = __commonJS({
         var ContextProvider = REACT_PROVIDER_TYPE;
         var Element2 = REACT_ELEMENT_TYPE;
         var ForwardRef = REACT_FORWARD_REF_TYPE;
-        var Fragment29 = REACT_FRAGMENT_TYPE;
+        var Fragment28 = REACT_FRAGMENT_TYPE;
         var Lazy = REACT_LAZY_TYPE;
         var Memo = REACT_MEMO_TYPE;
         var Portal = REACT_PORTAL_TYPE;
@@ -468061,7 +468079,7 @@ var require_react_is_development = __commonJS({
         exports2.ContextProvider = ContextProvider;
         exports2.Element = Element2;
         exports2.ForwardRef = ForwardRef;
-        exports2.Fragment = Fragment29;
+        exports2.Fragment = Fragment28;
         exports2.Lazy = Lazy;
         exports2.Memo = Memo;
         exports2.Portal = Portal;
@@ -501307,7 +501325,7 @@ __name(getPackageJson, "getPackageJson");
 // packages/cli/src/utils/version.ts
 async function getCliVersion() {
   const pkgJson = await getPackageJson();
-  return "1.0.0";
+  return "1.0.1";
 }
 __name(getCliVersion, "getCliVersion");
 
@@ -512976,7 +512994,7 @@ var formatDuration = /* @__PURE__ */ __name((milliseconds, options2) => {
 
 // packages/cli/src/generated/git-commit.ts
 init_esbuild_shims();
-var GIT_COMMIT_INFO = "ea0a15ca8";
+var GIT_COMMIT_INFO = "97ca1a8d7";
 
 // packages/cli/src/utils/systemInfo.ts
 async function getNpmVersion() {
@@ -552401,7 +552419,18 @@ function CompressionMessage({
       return t4("Compressing chat history");
     }
     switch (compressionStatus) {
-      case CompressionStatus.COMPRESSED:
+      case CompressionStatus.COMPRESSED: {
+        const reduction = originalTokens > 0 ? Math.round((originalTokens - newTokens) / originalTokens * 100) : 0;
+        if (reduction > 0) {
+          return t4(
+            "Chat history compressed: {{originalTokens}} \u2192 {{newTokens}} tokens ({{reduction}}% saved)",
+            {
+              originalTokens: String(originalTokens),
+              newTokens: String(newTokens),
+              reduction: String(reduction)
+            }
+          );
+        }
         return t4(
           "Chat history compressed from {{originalTokens}} to {{newTokens}} tokens.",
           {
@@ -552409,6 +552438,7 @@ function CompressionMessage({
             newTokens: String(newTokens)
           }
         );
+      }
       case CompressionStatus.COMPRESSION_FAILED_INFLATED_TOKEN_COUNT:
         if (originalTokens < 5e4) {
           return t4("Compression was not beneficial for this history size.");
@@ -552428,7 +552458,7 @@ function CompressionMessage({
   }, "getCompressionText");
   const text = getCompressionText();
   return /* @__PURE__ */ (0, import_jsx_runtime39.jsxs)(Box_default, { flexDirection: "row", children: [
-    /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(Box_default, { marginRight: 1, children: isPending ? /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(build_default, { type: "dots" }) : /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(Text3, { color: theme.text.accent, children: "\u2726" }) }),
+    /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(Box_default, { marginRight: 1, children: isPending ? /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(Text3, { color: theme.text.accent, children: /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(build_default, { type: "dots" }) }) : /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(Text3, { color: theme.text.accent, children: "\u2726" }) }),
     /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(Box_default, { children: /* @__PURE__ */ (0, import_jsx_runtime39.jsx)(
       Text3,
       {
@@ -569956,6 +569986,16 @@ var import_react136 = __toESM(require_react(), 1);
 init_esbuild_shims();
 init_i18n();
 var import_jsx_runtime129 = __toESM(require_jsx_runtime(), 1);
+function formatTokenCount2(tokens) {
+  if (tokens >= 1e6) {
+    return `${(tokens / 1e6).toFixed(1)}M`;
+  }
+  if (tokens >= 1e3) {
+    return `${(tokens / 1e3).toFixed(1)}k`;
+  }
+  return `${tokens}`;
+}
+__name(formatTokenCount2, "formatTokenCount");
 function formatPercentageUsed(percentage) {
   if (percentage > 1) {
     return ">100";
@@ -569974,16 +570014,39 @@ var ContextUsageDisplay = /* @__PURE__ */ __name(({
   const percentage = promptTokenCount / contextWindowSize;
   const percentageUsed = formatPercentageUsed(percentage);
   const isOverLimit = percentage > 1;
-  const label = terminalWidth < 100 ? t4("% used") : t4("% context used");
-  if (isOverLimit) {
-    return /* @__PURE__ */ (0, import_jsx_runtime129.jsx)(import_jsx_runtime129.Fragment, { children: /* @__PURE__ */ (0, import_jsx_runtime129.jsxs)(Text3, { color: theme.status.error, children: [
+  const tokenStr = formatTokenCount2(promptTokenCount);
+  if (terminalWidth < 80) {
+    const label2 = t4("% used");
+    return /* @__PURE__ */ (0, import_jsx_runtime129.jsxs)(Text3, { color: isOverLimit ? theme.status.error : theme.text.secondary, children: [
       percentageUsed,
-      label
-    ] }) });
+      label2
+    ] });
   }
-  return /* @__PURE__ */ (0, import_jsx_runtime129.jsxs)(Text3, { color: theme.text.secondary, children: [
+  if (terminalWidth < 120) {
+    const label2 = t4("% ctx");
+    return /* @__PURE__ */ (0, import_jsx_runtime129.jsxs)(Text3, { color: isOverLimit ? theme.status.error : theme.text.secondary, children: [
+      percentageUsed,
+      label2,
+      /* @__PURE__ */ (0, import_jsx_runtime129.jsxs)(Text3, { dimColor: true, children: [
+        " ",
+        "(",
+        tokenStr,
+        ")"
+      ] })
+    ] });
+  }
+  const label = t4("% context used");
+  return /* @__PURE__ */ (0, import_jsx_runtime129.jsxs)(Text3, { color: isOverLimit ? theme.status.error : theme.text.secondary, children: [
     percentageUsed,
-    label
+    label,
+    /* @__PURE__ */ (0, import_jsx_runtime129.jsxs)(Text3, { dimColor: true, children: [
+      " ",
+      "(",
+      tokenStr,
+      "/",
+      formatTokenCount2(contextWindowSize),
+      " tok)"
+    ] })
   ] });
 }, "ContextUsageDisplay");
 
@@ -577761,15 +577824,15 @@ var useLoadingIndicator = /* @__PURE__ */ __name((streamingState, customWittyPhr
   const [taskStartTokens, setTaskStartTokens] = (0, import_react170.useState)(0);
   const prevStreamingStateRef = (0, import_react170.useRef)(null);
   (0, import_react170.useEffect)(() => {
-    if (prevStreamingStateRef.current === "waiting_for_confirmation" /* WaitingForConfirmation */ && streamingState === "responding" /* Responding */) {
+    const prev = prevStreamingStateRef.current;
+    if (prev === "waiting_for_confirmation" /* WaitingForConfirmation */ && streamingState === "responding" /* Responding */) {
       setTimerResetKey((prevKey) => prevKey + 1);
       setRetainedElapsedTime(0);
-      setTaskStartTokens(currentCandidatesTokens ?? 0);
-    } else if (streamingState === "idle" /* Idle */ && prevStreamingStateRef.current === "responding" /* Responding */) {
+    } else if (streamingState === "idle" /* Idle */ && prev === "responding" /* Responding */) {
       setTimerResetKey((prevKey) => prevKey + 1);
       setRetainedElapsedTime(0);
       setTaskStartTokens(0);
-    } else if (streamingState === "responding" /* Responding */ && prevStreamingStateRef.current !== "responding" /* Responding */) {
+    } else if (streamingState === "responding" /* Responding */ && prev !== "responding" /* Responding */) {
       setTaskStartTokens(currentCandidatesTokens ?? 0);
     } else if (streamingState === "waiting_for_confirmation" /* WaitingForConfirmation */) {
       setRetainedElapsedTime(elapsedTimeFromTimer);
@@ -588845,7 +588908,7 @@ var VivekMindAgent = class {
   async initialize(args2) {
     this.clientCapabilities = args2.clientCapabilities;
     const authMethods = buildAuthMethods();
-    const version2 = "1.0.0";
+    const version2 = "1.0.1";
     return {
       protocolVersion: PROTOCOL_VERSION,
       agentInfo: {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vivekmind",
-  "version": "1.0.0",
+  "version": "1.0.1",
   "description": "Open-source terminal AI agent with full AWS Bedrock support. Bring your own keys.",
   "license": "Apache-2.0",
   "homepage": "https://code.vivekmind.com",
@@ -25,7 +25,7 @@
     "bundled"
   ],
   "config": {
-    "sandboxImageUri": "ghcr.io/lnxtanx/vivekmind-cli:1.0.0"
+    "sandboxImageUri": "ghcr.io/lnxtanx/vivekmind-cli:1.0.1"
   },
   "dependencies": {},
   "optionalDependencies": {