ur-agent 1.45.4 → 1.45.5

package/CHANGELOG.md

@@ -1,5 +1,20 @@
 # Changelog
 
+## 1.45.5
+
+- Bounded Ollama Cloud response-header and streaming phases to 120 seconds by
+  default while preserving the five-minute allowance for local Ollama models.
+  `API_TIMEOUT_MS` and per-request timeouts still take precedence.
+- Stopped deliberate Ollama Cloud stream deadlines from triggering the shared
+  non-streaming fallback and retry chain. Other providers and local Ollama
+  retain their existing fallback and retry behavior.
+- Extended the deterministic verifier to reject turns that end by promising an
+  immediate file change or command but emit no successful matching tool call.
+  Conditional plans and instructional prose remain unaffected.
+- Added focused timeout, retry-containment, intent-detection, and verifier
+  integration coverage; synchronized npm, IDE extension, static-site, user,
+  validation, and technical release metadata.
+
 ## 1.45.4
 
 - Added mandatory provider-first model selection for the first interactive run

package/README.md

@@ -636,8 +636,9 @@ the permission boundary matters.
   otherwise pause for permission approval, while user-input dialogs still ask.
 - `--dangerously-skip-permissions` should only be used inside disposable
   sandboxes.
-- The verifier checks for false completion claims, repeated tool-call loops,
-  empty assistant turns, and project gates.
+- The verifier checks for false completion claims, immediate-action promises
+  that end without the matching tool call, repeated tool-call loops, empty
+  assistant turns, and project gates.
 - Project gates can be configured in `.ur/verify.json`.
 - `ur test-first install` writes detected compile/test/lint commands into
   `.ur/verify.json` so mutating turns have command evidence before completion.

package/dist/cli.js

@@ -56931,6 +56931,7 @@ var init_kimiToolCalls = __esm(() => {
 var exports_ollama = {};
 __export(exports_ollama, {
   mergeToolCalls: () => mergeToolCalls,
+  isOllamaCloudModel: () => isOllamaCloudModel2,
   getOllamaRequestTimeoutMs: () => getOllamaRequestTimeoutMs,
   getEffectiveOllamaBaseUrl: () => getEffectiveOllamaBaseUrl,
   createOllamaURHQClient: () => createOllamaURHQClient
@@ -56986,7 +56987,7 @@ async function createNonStreamingRequest(params, options) {
   return ollamaResponseToURHQMessage(json2, params);
 }
 async function fetchOllamaChat(params, stream4, controller, options) {
-  const timeout = getOllamaRequestTimeoutMs(options);
+  const timeout = getOllamaRequestTimeoutMs(options, process.env, params.model);
   const timeoutId = timeout > 0 ? setTimeout(() => controller.abort(), timeout) : undefined;
   try {
     const capabilities = await getOllamaModelCapabilities(params.model, controller.signal);
@@ -57064,7 +57065,7 @@ function createLinkedAbortController(options) {
   signal.addEventListener("abort", () => controller.abort(), { once: true });
   return controller;
 }
-function getOllamaRequestTimeoutMs(options, env4 = process.env) {
+function getOllamaRequestTimeoutMs(options, env4 = process.env, model) {
   if (options?.timeoutMs !== undefined || options?.timeout !== undefined) {
     return options.timeoutMs ?? options.timeout ?? DEFAULT_OLLAMA_REQUEST_TIMEOUT_MS;
   }
@@ -57072,7 +57073,16 @@ function getOllamaRequestTimeoutMs(options, env4 = process.env) {
   if (override > 0) {
     return override;
   }
-  return isTruthyEnv(env4.UR_CODE_REMOTE) ? REMOTE_OLLAMA_REQUEST_TIMEOUT_MS : DEFAULT_OLLAMA_REQUEST_TIMEOUT_MS;
+  if (isTruthyEnv(env4.UR_CODE_REMOTE)) {
+    return REMOTE_OLLAMA_REQUEST_TIMEOUT_MS;
+  }
+  if (isOllamaCloudModel2(model)) {
+    return CLOUD_OLLAMA_REQUEST_TIMEOUT_MS;
+  }
+  return DEFAULT_OLLAMA_REQUEST_TIMEOUT_MS;
+}
+function isOllamaCloudModel2(model) {
+  return model?.trim().toLowerCase().endsWith(":cloud") ?? false;
 }
 function isTruthyEnv(value) {
   if (!value) {
@@ -57473,7 +57483,7 @@ async function* streamURHQEvents(response, params, controller, requestId, option
     }
     return events;
   };
-  for await (const chunk of readOllamaChunks(response, controller, getOllamaRequestTimeoutMs(options), options)) {
+  for await (const chunk of readOllamaChunks(response, controller, getOllamaRequestTimeoutMs(options, process.env, params.model), options)) {
     if (chunk.error) {
       throw new Error(chunk.error);
     }
@@ -57920,7 +57930,7 @@ function parseToolInput(input) {
   }
   return parsed ?? {};
 }
-var DEFAULT_OLLAMA_REQUEST_TIMEOUT_MS = 300000, REMOTE_OLLAMA_REQUEST_TIMEOUT_MS = 120000, OLLAMA_GATEWAY_TIMEOUT_MESSAGE = "Ollama gateway timed out while waiting for the model to respond. Check the selected Ollama endpoint or increase API_TIMEOUT_MS if the model needs more time.", ollamaModelCapabilitiesCache, ollamaBaseUrlOverride, warnedToolsUnsupportedModels, TEXT_TOOL_CALL_HINT;
+var DEFAULT_OLLAMA_REQUEST_TIMEOUT_MS = 300000, REMOTE_OLLAMA_REQUEST_TIMEOUT_MS = 120000, CLOUD_OLLAMA_REQUEST_TIMEOUT_MS = 120000, OLLAMA_GATEWAY_TIMEOUT_MESSAGE = "Ollama gateway timed out while waiting for the model to respond. Check the selected Ollama endpoint or increase API_TIMEOUT_MS if the model needs more time.", ollamaModelCapabilitiesCache, ollamaBaseUrlOverride, warnedToolsUnsupportedModels, TEXT_TOOL_CALL_HINT;
 var init_ollama = __esm(() => {
   init_urhq_sdk();
   init_ollamaModels();
@@ -73087,7 +73097,7 @@ var init_auth = __esm(() => {
 
 // src/utils/userAgent.ts
 function getURCodeUserAgent() {
-  return `ur/${"1.45.4"}`;
+  return `ur/${"1.45.5"}`;
 }
 
 // src/utils/workloadContext.ts
@@ -73109,7 +73119,7 @@ function getUserAgent() {
   const clientApp = process.env.UR_AGENT_SDK_CLIENT_APP ? `, client-app/${process.env.UR_AGENT_SDK_CLIENT_APP}` : "";
   const workload = getWorkload();
   const workloadSuffix = workload ? `, workload/${workload}` : "";
-  return `ur-cli/${"1.45.4"} (${process.env.USER_TYPE}, ${process.env.UR_CODE_ENTRYPOINT ?? "cli"}${agentSdkVersion}${clientApp}${workloadSuffix})`;
+  return `ur-cli/${"1.45.5"} (${process.env.USER_TYPE}, ${process.env.UR_CODE_ENTRYPOINT ?? "cli"}${agentSdkVersion}${clientApp}${workloadSuffix})`;
 }
 function getMCPUserAgent() {
   const parts = [];
@@ -73123,7 +73133,7 @@ function getMCPUserAgent() {
     parts.push(`client-app/${process.env.UR_AGENT_SDK_CLIENT_APP}`);
   }
   const suffix = parts.length > 0 ? ` (${parts.join(", ")})` : "";
-  return `ur/${"1.45.4"}${suffix}`;
+  return `ur/${"1.45.5"}${suffix}`;
 }
 function getWebFetchUserAgent() {
   return `UR-User (${getURCodeUserAgent()})`;
@@ -73261,7 +73271,7 @@ var init_user = __esm(() => {
       deviceId,
       sessionId: getSessionId(),
       email: getEmail(),
-      appVersion: "1.45.4",
+      appVersion: "1.45.5",
       platform: getHostPlatformForAnalytics(),
       organizationUuid,
       accountUuid,
@@ -79836,7 +79846,7 @@ var init_metadata = __esm(() => {
   COMPOUND_OPERATOR_REGEX = /\s*(?:&&|\|\||[;|])\s*/;
   WHITESPACE_REGEX = /\s+/;
   getVersionBase = memoize_default(() => {
-    const match = "1.45.4".match(/^\d+\.\d+\.\d+(?:-[a-z]+)?/);
+    const match = "1.45.5".match(/^\d+\.\d+\.\d+(?:-[a-z]+)?/);
     return match ? match[0] : undefined;
   });
   buildEnvContext = memoize_default(async () => {
@@ -79876,7 +79886,7 @@ var init_metadata = __esm(() => {
       isGithubAction: isEnvTruthy(process.env.GITHUB_ACTIONS),
       isURCodeAction: isEnvTruthy(process.env.UR_CODE_ACTION),
       isURAiAuth: isURAISubscriber(),
-      version: "1.45.4",
+      version: "1.45.5",
       versionBase: getVersionBase(),
       buildTime: "",
       deploymentEnvironment: env2.detectDeploymentEnvironment(),
@@ -80546,7 +80556,7 @@ function initialize1PEventLogging() {
   const platform2 = getPlatform();
   const attributes = {
     [import_semantic_conventions4.ATTR_SERVICE_NAME]: "ur",
-    [import_semantic_conventions4.ATTR_SERVICE_VERSION]: "1.45.4"
+    [import_semantic_conventions4.ATTR_SERVICE_VERSION]: "1.45.5"
   };
   if (platform2 === "wsl") {
     const wslVersion = getWslVersion();
@@ -80573,7 +80583,7 @@ function initialize1PEventLogging() {
       })
     ]
   });
-  firstPartyEventLogger = firstPartyEventLoggerProvider.getLogger("com.urhq.ur.events", "1.45.4");
+  firstPartyEventLogger = firstPartyEventLoggerProvider.getLogger("com.urhq.ur.events", "1.45.5");
 }
 async function reinitialize1PEventLoggingIfConfigChanged() {
   if (!is1PEventLoggingEnabled() || !firstPartyEventLoggerProvider) {
@@ -86498,7 +86508,7 @@ function formatAgentTrendReport(report = buildAgentTrendReport()) {
 function formatA2AAgentCard(options = {}, pretty = true) {
   return JSON.stringify(buildA2AAgentCard(options), null, pretty ? 2 : 0);
 }
-var urVersion = "1.45.4", coverage, priorityRoadmap;
+var urVersion = "1.45.5", coverage, priorityRoadmap;
 var init_trends = __esm(() => {
   coverage = [
     {
@@ -88364,7 +88374,7 @@ function getAttributionHeader(fingerprint) {
   if (!isAttributionHeaderEnabled()) {
     return "";
   }
-  const version2 = `${"1.45.4"}.${fingerprint}`;
+  const version2 = `${"1.45.5"}.${fingerprint}`;
   const entrypoint = process.env.UR_CODE_ENTRYPOINT ?? "unknown";
   const cch = "";
   const workload = getWorkload();
@@ -146550,6 +146560,25 @@ var init_projectQuality = __esm(() => {
 });
 
 // src/services/verifier/doneDetector.ts
+function detectImmediateIntent(text) {
+  const tail = text.trim().slice(-600);
+  const clauses = tail.split(/(?:\n+|(?<=[.!?])\s+)/).map((clause2) => clause2.trim()).filter(Boolean);
+  const clause = clauses.at(-1) ?? "";
+  if (!clause || /^(?:if|when|once|after|before)\b/i.test(clause) || /\b(?:if|when|once|after|unless|until)\b.+$/i.test(clause)) {
+    return null;
+  }
+  const lead = "(?:now[, :]*)?(?:let me|i(?:'ll| will| am going to|'m going to))";
+  if (new RegExp(`\\b${lead}\\s+(?:now\\s+)?${WRITE_INTENT_VERBS}\\b`, "i").test(clause)) {
+    return "write_intent";
+  }
+  if (new RegExp(`\\b${lead}\\s+(?:now\\s+)?${RUN_INTENT_VERBS}\\b`, "i").test(clause)) {
+    return "run_intent";
+  }
+  if (new RegExp(`^${WRITE_INTENT_GERUNDS}\\b.*\\bnow[.!]?$`, "i").test(clause)) {
+    return "write_intent";
+  }
+  return null;
+}
 function detectDoneClaim(text) {
   if (!text)
     return null;
@@ -146569,7 +146598,7 @@ function detectDoneClaim(text) {
     if (pattern.test(text))
       return "generic_done";
   }
-  return null;
+  return detectImmediateIntent(text);
 }
 function evaluateDoneGate(claim, hasMutatingEffect, ranBash) {
   if (claim === "write_claim" || claim === "edit_claim" || claim === "delete_claim") {
@@ -146582,6 +146611,16 @@ function evaluateDoneGate(claim, hasMutatingEffect, ranBash) {
       reminder: "You claimed to have created, edited, or deleted files this turn but no Write / Edit / NotebookEdit / Bash tool call returned successfully. Make the actual tool call now, or correct the statement before continuing."
     };
   }
+  if (claim === "write_intent") {
+    if (hasMutatingEffect)
+      return { ok: true };
+    return {
+      ok: false,
+      claim,
+      reason: "promised file action ended without a mutating tool call",
+      reminder: "You ended by saying you were about to create, edit, or fix files, but no Write / Edit / NotebookEdit / Bash tool call returned successfully. Make the actual tool call now, or explain why you cannot continue."
+    };
+  }
   if (claim === "run_claim") {
     if (ranBash)
       return { ok: true };
@@ -146592,6 +146631,16 @@ function evaluateDoneGate(claim, hasMutatingEffect, ranBash) {
       reminder: "You claimed to have run a command this turn but no Bash tool call returned successfully. Run the command now or correct the statement."
     };
   }
+  if (claim === "run_intent") {
+    if (ranBash)
+      return { ok: true };
+    return {
+      ok: false,
+      claim,
+      reason: "promised command ended without a successful Bash call",
+      reminder: "You ended by saying you were about to run a command, but no Bash tool call returned successfully. Run the command now, or explain why you cannot continue."
+    };
+  }
   if (hasMutatingEffect || ranBash)
     return { ok: true };
   return {
@@ -146601,7 +146650,7 @@ function evaluateDoneGate(claim, hasMutatingEffect, ranBash) {
     reminder: "You declared the task complete but this turn made no Write / Edit / Bash / NotebookEdit tool call. If the task required no edits, say so explicitly. Otherwise, make the tool call now."
   };
 }
-var DONE_PATTERNS;
+var DONE_PATTERNS, WRITE_INTENT_VERBS = "(?:create|write|edit|update|change|modify|patch|fix|remove|delete|rewrite|implement)", RUN_INTENT_VERBS = "(?:run|execute|test|build)", WRITE_INTENT_GERUNDS = "(?:creating|writing|editing|updating|changing|modifying|patching|fixing|removing|deleting|rewriting|implementing)";
 var init_doneDetector = __esm(() => {
   DONE_PATTERNS = [
     /\b(?:i|i've|i have)\s+(?:created|added|written|wrote|made)\b/i,
@@ -196545,7 +196594,7 @@ function getTelemetryAttributes() {
     attributes["session.id"] = sessionId;
   }
   if (shouldIncludeAttribute("OTEL_METRICS_INCLUDE_VERSION")) {
-    attributes["app.version"] = "1.45.4";
+    attributes["app.version"] = "1.45.5";
   }
   const oauthAccount = getOauthAccountInfo();
   if (oauthAccount) {
@@ -232026,7 +232075,7 @@ function getInstallationEnv() {
   return;
 }
 function getURCodeVersion() {
-  return "1.45.4";
+  return "1.45.5";
 }
 async function getInstalledVSCodeExtensionVersion(command) {
   const { stdout } = await execFileNoThrow(command, ["--list-extensions", "--show-versions"], {
@@ -239408,7 +239457,7 @@ async function setupSdkMcpClients(sdkMcpConfigs, sendMcpMessage) {
     const client2 = new Client({
       name: "ur",
       title: "UR",
-      version: "1.45.4",
+      version: "1.45.5",
       description: "UR-Nexus autonomous engineering workflow engine",
       websiteUrl: PRODUCT_URL
     }, {
@@ -239771,7 +239820,7 @@ var init_client5 = __esm(() => {
       const client2 = new Client({
         name: "ur",
         title: "UR",
-        version: "1.45.4",
+        version: "1.45.5",
         description: "UR-Nexus autonomous engineering workflow engine",
         websiteUrl: PRODUCT_URL
       }, {
@@ -246808,9 +246857,9 @@ async function assertMinVersion() {
   if (false) {}
   try {
     const versionConfig = await getDynamicConfig_BLOCKS_ON_INIT("tengu_version_config", { minVersion: "0.0.0" });
-    if (versionConfig.minVersion && lt("1.45.4", versionConfig.minVersion)) {
+    if (versionConfig.minVersion && lt("1.45.5", versionConfig.minVersion)) {
       console.error(`
-It looks like your version of UR (${"1.45.4"}) needs an update.
+It looks like your version of UR (${"1.45.5"}) needs an update.
 A newer version (${versionConfig.minVersion} or higher) is required to continue.
 
 To update, please run:
@@ -247026,7 +247075,7 @@ async function installGlobalPackage(specificVersion) {
     logError2(new AutoUpdaterError("Another process is currently installing an update"));
     logEvent("tengu_auto_updater_lock_contention", {
       pid: process.pid,
-      currentVersion: "1.45.4"
+      currentVersion: "1.45.5"
     });
     return "in_progress";
   }
@@ -247035,7 +247084,7 @@ async function installGlobalPackage(specificVersion) {
     if (!env2.isRunningWithBun() && env2.isNpmFromWindowsPath()) {
       logError2(new Error("Windows NPM detected in WSL environment"));
       logEvent("tengu_auto_updater_windows_npm_in_wsl", {
-        currentVersion: "1.45.4"
+        currentVersion: "1.45.5"
       });
       console.error(`
 Error: Windows NPM detected in WSL
@@ -247570,7 +247619,7 @@ function detectLinuxGlobPatternWarnings() {
 }
 async function getDoctorDiagnostic() {
   const installationType = await getCurrentInstallationType();
-  const version2 = typeof MACRO !== "undefined" ? "1.45.4" : "unknown";
+  const version2 = typeof MACRO !== "undefined" ? "1.45.5" : "unknown";
   const installationPath = await getInstallationPath();
   const invokedBinary = getInvokedBinary();
   const multipleInstallations = await detectMultipleInstallations();
@@ -248505,8 +248554,8 @@ async function updateLatest(channelOrVersion, forceReinstall = false) {
     const maxVersion = await getMaxVersion();
     if (maxVersion && gt(version2, maxVersion)) {
       logForDebugging(`Native installer: maxVersion ${maxVersion} is set, capping update from ${version2} to ${maxVersion}`);
-      if (gte("1.45.4", maxVersion)) {
-        logForDebugging(`Native installer: current version ${"1.45.4"} is already at or above maxVersion ${maxVersion}, skipping update`);
+      if (gte("1.45.5", maxVersion)) {
+        logForDebugging(`Native installer: current version ${"1.45.5"} is already at or above maxVersion ${maxVersion}, skipping update`);
         logEvent("tengu_native_update_skipped_max_version", {
           latency_ms: Date.now() - startTime,
           max_version: maxVersion,
@@ -248517,7 +248566,7 @@ async function updateLatest(channelOrVersion, forceReinstall = false) {
       version2 = maxVersion;
     }
   }
-  if (!forceReinstall && version2 === "1.45.4" && await versionIsAvailable(version2) && await isPossibleURBinary(executablePath)) {
+  if (!forceReinstall && version2 === "1.45.5" && await versionIsAvailable(version2) && await isPossibleURBinary(executablePath)) {
     logForDebugging(`Found ${version2} at ${executablePath}, skipping install`);
     logEvent("tengu_native_update_complete", {
       latency_ms: Date.now() - startTime,
@@ -346377,7 +346426,7 @@ function Feedback({
       platform: env2.platform,
       gitRepo: envInfo.isGit,
       terminal: env2.terminal,
-      version: "1.45.4",
+      version: "1.45.5",
       transcript: normalizeMessagesForAPI(messages),
       errors: sanitizedErrors,
       lastApiRequest: getLastAPIRequest(),
@@ -346569,7 +346618,7 @@ function Feedback({
                       ", ",
                       env2.terminal,
                       ", v",
-                      "1.45.4"
+                      "1.45.5"
                     ]
                   }, undefined, true, undefined, this)
                 ]
@@ -346675,7 +346724,7 @@ ${sanitizedDescription}
 ` + `**Environment Info**
 ` + `- Platform: ${env2.platform}
 ` + `- Terminal: ${env2.terminal}
-` + `- Version: ${"1.45.4"}
+` + `- Version: ${"1.45.5"}
 ` + `- Feedback ID: ${feedbackId}
 ` + `
 **Errors**
@@ -349786,7 +349835,7 @@ function buildPrimarySection() {
   }, undefined, false, undefined, this);
   return [{
     label: "Version",
-    value: "1.45.4"
+    value: "1.45.5"
   }, {
     label: "Session name",
     value: nameValue
@@ -353116,7 +353165,7 @@ function Config({
         }
       }, undefined, false, undefined, this)
     }, undefined, false, undefined, this) : showSubmenu === "ChannelDowngrade" ? /* @__PURE__ */ jsx_dev_runtime178.jsxDEV(ChannelDowngradeDialog, {
-      currentVersion: "1.45.4",
+      currentVersion: "1.45.5",
       onChoice: (choice) => {
         setShowSubmenu(null);
         setTabsHidden(false);
@@ -353128,7 +353177,7 @@ function Config({
           autoUpdatesChannel: "stable"
         };
         if (choice === "stay") {
-          newSettings.minimumVersion = "1.45.4";
+          newSettings.minimumVersion = "1.45.5";
         }
         updateSettingsForSource("userSettings", newSettings);
         setSettingsData((prev_27) => ({
@@ -361198,7 +361247,7 @@ function HelpV2(t0) {
   let t6;
   if ($3[31] !== tabs) {
     t6 = /* @__PURE__ */ jsx_dev_runtime205.jsxDEV(Tabs, {
-      title: `UR v${"1.45.4"}`,
+      title: `UR v${"1.45.5"}`,
       color: "professionalBlue",
       defaultTab: "general",
       children: tabs
@@ -361972,7 +362021,7 @@ function buildToolUseContext(tools, readFileStateCache) {
 async function handleInitialize(options2) {
   return {
     name: "UR",
-    version: "1.45.4",
+    version: "1.45.5",
     protocolVersion: "0.1.0",
     workspaceRoot: options2.cwd,
     capabilities: {
@@ -382114,7 +382163,7 @@ function getAllReleaseNotes(changelogContent = getStoredChangelogFromMemory()) {
     return [];
   }
 }
-async function checkForReleaseNotes(lastSeenVersion, currentVersion = "1.45.4") {
+async function checkForReleaseNotes(lastSeenVersion, currentVersion = "1.45.5") {
   if (process.env.USER_TYPE === "ant") {
     const changelog = "";
     if (changelog) {
@@ -382141,7 +382190,7 @@ async function checkForReleaseNotes(lastSeenVersion, currentVersion = "1.45.4")
     releaseNotes
   };
 }
-function checkForReleaseNotesSync(lastSeenVersion, currentVersion = "1.45.4") {
+function checkForReleaseNotesSync(lastSeenVersion, currentVersion = "1.45.5") {
   if (process.env.USER_TYPE === "ant") {
     const changelog = "";
     if (changelog) {
@@ -385233,7 +385282,7 @@ function getRecentActivitySync() {
   return cachedActivity;
 }
 function getLogoDisplayData() {
-  const version2 = process.env.DEMO_VERSION ?? "1.45.4";
+  const version2 = process.env.DEMO_VERSION ?? "1.45.5";
   const serverUrl = getDirectConnectServerUrl();
   const displayPath = process.env.DEMO_VERSION ? "/code/ur" : getDisplayPath(getCwd());
   const cwd2 = serverUrl ? `${displayPath} in ${serverUrl.replace(/^https?:\/\//, "")}` : displayPath;
@@ -386117,7 +386166,7 @@ function LogoV2() {
   if ($3[2] === Symbol.for("react.memo_cache_sentinel")) {
     t2 = () => {
       const currentConfig2 = getGlobalConfig();
-      if (currentConfig2.lastReleaseNotesSeen === "1.45.4") {
+      if (currentConfig2.lastReleaseNotesSeen === "1.45.5") {
         return;
       }
       saveGlobalConfig(_temp327);
@@ -386802,12 +386851,12 @@ function LogoV2() {
   return t41;
 }
 function _temp327(current) {
-  if (current.lastReleaseNotesSeen === "1.45.4") {
+  if (current.lastReleaseNotesSeen === "1.45.5") {
     return current;
   }
   return {
     ...current,
-    lastReleaseNotesSeen: "1.45.4"
+    lastReleaseNotesSeen: "1.45.5"
   };
 }
 function _temp243(s_0) {
@@ -604850,7 +604899,7 @@ async function captureMemoryDiagnostics(trigger2, dumpNumber = 0) {
     smapsRollup,
     platform: process.platform,
     nodeVersion: process.version,
-    ccVersion: "1.45.4"
+    ccVersion: "1.45.5"
   };
 }
 async function performHeapDump(trigger2 = "manual", dumpNumber = 0) {
@@ -605436,7 +605485,7 @@ var init_bridge_kick = __esm(() => {
 var call143 = async () => {
   return {
     type: "text",
-    value: "1.45.4"
+    value: "1.45.5"
   };
 }, version2, version_default;
 var init_version = __esm(() => {
@@ -616546,7 +616595,7 @@ function generateHtmlReport(data, insights) {
 </html>`;
 }
 function buildExportData(data, insights, facets, remoteStats) {
-  const version3 = typeof MACRO !== "undefined" ? "1.45.4" : "unknown";
+  const version3 = typeof MACRO !== "undefined" ? "1.45.5" : "unknown";
   const remote_hosts_collected = remoteStats?.hosts.filter((h2) => h2.sessionCount > 0).map((h2) => h2.name);
   const facets_summary = {
     total: facets.size,
@@ -620875,7 +620924,7 @@ var init_sessionStorage = __esm(() => {
   init_settings2();
   init_slowOperations();
   init_uuid();
-  VERSION5 = typeof MACRO !== "undefined" ? "1.45.4" : "unknown";
+  VERSION5 = typeof MACRO !== "undefined" ? "1.45.5" : "unknown";
   MAX_TOMBSTONE_REWRITE_BYTES = 50 * 1024 * 1024;
   SKIP_FIRST_PROMPT_PATTERN = /^(?:\s*<[a-z][\w-]*[\s>]|\[Request interrupted by user[^\]]*\])/;
   EPHEMERAL_PROGRESS_TYPES = new Set([
@@ -622080,7 +622129,7 @@ var init_filesystem = __esm(() => {
   });
   getBundledSkillsRoot = memoize_default(function getBundledSkillsRoot2() {
     const nonce = randomBytes18(16).toString("hex");
-    return join211(getURTempDir(), "bundled-skills", "1.45.4", nonce);
+    return join211(getURTempDir(), "bundled-skills", "1.45.5", nonce);
   });
   getResolvedWorkingDirPaths = memoize_default(getPathsForPermissionCheck);
 });
@@ -628375,7 +628424,7 @@ function computeFingerprint(messageText2, version3) {
 }
 function computeFingerprintFromMessages(messages) {
   const firstMessageText = extractFirstMessageText(messages);
-  return computeFingerprint(firstMessageText, "1.45.4");
+  return computeFingerprint(firstMessageText, "1.45.5");
 }
 var FINGERPRINT_SALT = "59cf53e54c78";
 var init_fingerprint = () => {};
@@ -628763,14 +628812,20 @@ function shouldDeferLspTool(tool) {
   const status2 = getInitializationStatus();
   return status2.status === "pending" || status2.status === "not-started";
 }
-function getNonstreamingFallbackTimeoutMs() {
-  const override = parseInt(process.env.API_TIMEOUT_MS || "", 10);
+function isOllamaCloudRuntime(model, provider = getAPIProvider()) {
+  return provider === "ollama" && model.trim().toLowerCase().endsWith(":cloud");
+}
+function getNonstreamingFallbackTimeoutMs(model, env4 = process.env, provider = getAPIProvider()) {
+  const override = parseInt(env4.API_TIMEOUT_MS || "", 10);
   if (override)
     return override;
-  return isEnvTruthy(process.env.UR_CODE_REMOTE) ? 120000 : 300000;
+  return isEnvTruthy(env4.UR_CODE_REMOTE) || isOllamaCloudRuntime(model, provider) ? 120000 : 300000;
+}
+function shouldSkipOllamaNonStreamingFallback(error40, model, provider = getAPIProvider()) {
+  return isOllamaCloudRuntime(model, provider) && error40 instanceof APIConnectionTimeoutError && error40.message === "Ollama stream timed out";
 }
 async function* executeNonStreamingRequest(clientOptions, retryOptions, paramsFromContext, onAttempt, captureRequest, originatingRequestId) {
-  const fallbackTimeoutMs = getNonstreamingFallbackTimeoutMs();
+  const fallbackTimeoutMs = getNonstreamingFallbackTimeoutMs(clientOptions.model);
   const generator = withRetry(() => getURHQClient({
     maxRetries: 0,
     model: clientOptions.model,
@@ -628804,6 +628859,7 @@ async function* executeNonStreamingRequest(clientOptions, retryOptions, paramsFr
       throw err2;
     }
   }, {
+    ...isOllamaCloudRuntime(retryOptions.model) ? { maxRetries: 0 } : {},
     model: retryOptions.model,
     fallbackModel: retryOptions.fallbackModel,
     thinkingConfig: retryOptions.thinkingConfig,
@@ -629283,6 +629339,7 @@ ${deferredToolList}
       streamResponse = result.response;
       return result.data;
     }, {
+      ...isOllamaCloudRuntime(options2.model) ? { maxRetries: 0 } : {},
       model: options2.model,
       fallbackModel: options2.fallbackModel,
       thinkingConfig,
@@ -629627,6 +629684,9 @@ ${deferredToolList}
           throw new APIConnectionTimeoutError({ message: "Request timed out" });
         }
       }
+      if (shouldSkipOllamaNonStreamingFallback(streamingError, options2.model)) {
+        throw streamingError;
+      }
       const disableFallback = isEnvTruthy(process.env.UR_CODE_DISABLE_NONSTREAMING_FALLBACK) || getFeatureValue_CACHED_MAY_BE_STALE("tengu_disable_streaming_to_non_streaming_fallback", false);
       if (disableFallback) {
         logForDebugging(`Error streaming (non-streaming fallback disabled): ${errorMessage2(streamingError)}`, { level: "error" });
@@ -630249,7 +630309,7 @@ async function sideQuery(opts) {
     betas.push(STRUCTURED_OUTPUTS_BETA_HEADER);
   }
   const messageText2 = extractFirstUserMessageText(messages);
-  const fingerprint2 = computeFingerprint(messageText2, "1.45.4");
+  const fingerprint2 = computeFingerprint(messageText2, "1.45.5");
   const attributionHeader = getAttributionHeader(fingerprint2);
   const systemBlocks = [
     attributionHeader ? { type: "text", text: attributionHeader } : null,
@@ -634986,7 +635046,7 @@ function buildSystemInitMessage(inputs) {
     slash_commands: inputs.commands.filter((c4) => c4.userInvocable !== false).map((c4) => c4.name),
     apiKeySource: getURHQApiKeyWithSource().source,
     betas: getSdkBetas(),
-    ur_version: "1.45.4",
+    ur_version: "1.45.5",
     output_style: outputStyle2,
     agents: inputs.agents.map((agent) => agent.agentType),
     skills: inputs.skills.filter((s) => s.userInvocable !== false).map((skill2) => skill2.name),
@@ -648846,7 +648906,7 @@ var init_useVoiceEnabled = __esm(() => {
 function getSemverPart(version3) {
   return `${import_semver13.major(version3, { loose: true })}.${import_semver13.minor(version3, { loose: true })}.${import_semver13.patch(version3, { loose: true })}`;
 }
-function useUpdateNotification(updatedVersion, initialVersion = "1.45.4") {
+function useUpdateNotification(updatedVersion, initialVersion = "1.45.5") {
   const [lastNotifiedSemver, setLastNotifiedSemver] = import_react228.useState(() => getSemverPart(initialVersion));
   if (!updatedVersion) {
     return null;
@@ -648895,7 +648955,7 @@ function AutoUpdater({
       return;
     }
     if (false) {}
-    const currentVersion = "1.45.4";
+    const currentVersion = "1.45.5";
     const channel = getInitialSettings()?.autoUpdatesChannel ?? "latest";
     let latestVersion = await getLatestVersion(channel);
     const isDisabled = isAutoUpdaterDisabled();
@@ -649124,12 +649184,12 @@ function NativeAutoUpdater({
     logEvent("tengu_native_auto_updater_start", {});
     try {
       const maxVersion = await getMaxVersion();
-      if (maxVersion && gt("1.45.4", maxVersion)) {
+      if (maxVersion && gt("1.45.5", maxVersion)) {
         const msg = await getMaxVersionMessage();
         setMaxVersionIssue(msg ?? "affects your version");
       }
       const result = await installLatest(channel);
-      const currentVersion = "1.45.4";
+      const currentVersion = "1.45.5";
       const latencyMs = Date.now() - startTime;
       if (result.lockFailed) {
         logEvent("tengu_native_auto_updater_lock_contention", {
@@ -649266,17 +649326,17 @@ function PackageManagerAutoUpdater(t0) {
       const maxVersion = await getMaxVersion();
       if (maxVersion && latest && gt(latest, maxVersion)) {
         logForDebugging(`PackageManagerAutoUpdater: maxVersion ${maxVersion} is set, capping update from ${latest} to ${maxVersion}`);
-        if (gte("1.45.4", maxVersion)) {
-          logForDebugging(`PackageManagerAutoUpdater: current version ${"1.45.4"} is already at or above maxVersion ${maxVersion}, skipping update`);
+        if (gte("1.45.5", maxVersion)) {
+          logForDebugging(`PackageManagerAutoUpdater: current version ${"1.45.5"} is already at or above maxVersion ${maxVersion}, skipping update`);
           setUpdateAvailable(false);
           return;
         }
         latest = maxVersion;
       }
-      const hasUpdate = latest && !gte("1.45.4", latest) && !shouldSkipVersion(latest);
+      const hasUpdate = latest && !gte("1.45.5", latest) && !shouldSkipVersion(latest);
       setUpdateAvailable(!!hasUpdate);
       if (hasUpdate) {
-        logForDebugging(`PackageManagerAutoUpdater: Update available ${"1.45.4"} -> ${latest}`);
+        logForDebugging(`PackageManagerAutoUpdater: Update available ${"1.45.5"} -> ${latest}`);
       }
     };
     $3[0] = t1;
@@ -649310,7 +649370,7 @@ function PackageManagerAutoUpdater(t0) {
       wrap: "truncate",
       children: [
         "currentVersion: ",
-        "1.45.4"
+        "1.45.5"
       ]
     }, undefined, true, undefined, this);
     $3[3] = verbose;
@@ -660007,7 +660067,7 @@ function buildStatusLineCommandInput(permissionMode, exceeds200kTokens, settings
       project_dir: getOriginalCwd(),
       added_dirs: addedDirs
     },
-    version: "1.45.4",
+    version: "1.45.5",
     output_style: {
       name: outputStyleName
     },
@@ -660090,7 +660150,7 @@ function StatusLineInner({
   const taskValues = Object.values(tasks2);
   const taskRunningCount = taskValues.filter((task2) => task2.status === "running" || task2.status === "pending").length;
   const defaultStatusLineText = buildDefaultStatusBar({
-    version: "1.45.4",
+    version: "1.45.5",
     providerLabel: providerRuntime.providerLabel,
     authMode: providerRuntime.authLabel,
     model: providerRuntime.model ?? renderModelName(mainLoopModel),
@@ -672083,7 +672143,7 @@ async function submitTranscriptShare(messages, trigger2, appearanceId) {
     } catch {}
     const data = {
       trigger: trigger2,
-      version: "1.45.4",
+      version: "1.45.5",
       platform: process.platform,
       transcript,
       subagentTranscripts: Object.keys(subagentTranscripts).length > 0 ? subagentTranscripts : undefined,
@@ -684369,7 +684429,7 @@ function WelcomeV2() {
             dimColor: true,
             children: [
               "v",
-              "1.45.4"
+              "1.45.5"
             ]
           }, undefined, true, undefined, this)
         ]
@@ -685629,7 +685689,7 @@ function completeOnboarding() {
   saveGlobalConfig((current) => ({
     ...current,
     hasCompletedOnboarding: true,
-    lastOnboardingVersion: "1.45.4"
+    lastOnboardingVersion: "1.45.5"
   }));
 }
 function showDialog(root2, renderer) {
@@ -690651,7 +690711,7 @@ function appendToLog(path24, message) {
     cwd: getFsImplementation().cwd(),
     userType: process.env.USER_TYPE,
     sessionId: getSessionId(),
-    version: "1.45.4"
+    version: "1.45.5"
   };
   getLogWriter(path24).write(messageWithTimestamp);
 }
@@ -694810,8 +694870,8 @@ async function getEnvLessBridgeConfig() {
 }
 async function checkEnvLessBridgeMinVersion() {
   const cfg = await getEnvLessBridgeConfig();
-  if (cfg.min_version && lt("1.45.4", cfg.min_version)) {
-    return `Your version of UR (${"1.45.4"}) is too old for Remote Control.
+  if (cfg.min_version && lt("1.45.5", cfg.min_version)) {
+    return `Your version of UR (${"1.45.5"}) is too old for Remote Control.
 Version ${cfg.min_version} or higher is required. Run \`ur update\` to update.`;
   }
   return null;
@@ -695285,7 +695345,7 @@ async function initBridgeCore(params) {
   const rawApi = createBridgeApiClient({
     baseUrl,
     getAccessToken,
-    runnerVersion: "1.45.4",
+    runnerVersion: "1.45.5",
     onDebug: logForDebugging,
     onAuth401,
     getTrustedDeviceToken
@@ -700967,7 +701027,7 @@ async function startMCPServer(cwd3, debug2, verbose) {
   setCwd(cwd3);
   const server2 = new Server({
     name: "ur/tengu",
-    version: "1.45.4"
+    version: "1.45.5"
   }, {
     capabilities: {
       tools: {}
@@ -703009,7 +703069,7 @@ async function update() {
   logEvent("tengu_update_check", {});
   const diagnostic = await getDoctorDiagnostic();
   const result = await checkUpgradeStatus({
-    currentVersion: "1.45.4",
+    currentVersion: "1.45.5",
     packageName: UR_AGENT_PACKAGE_NAME,
     installationType: diagnostic.installationType,
     latestVersion: () => getLatestNpmPackageVersion(UR_AGENT_PACKAGE_NAME)
@@ -704320,7 +704380,7 @@ ${customInstructions}` : customInstructions;
       }
     }
     logForDiagnosticsNoPII("info", "started", {
-      version: "1.45.4",
+      version: "1.45.5",
       is_native_binary: isInBundledMode()
     });
     registerCleanup(async () => {
@@ -705106,7 +705166,7 @@ Usage: ur --remote "your task description"`, () => gracefulShutdown(1));
         pendingHookMessages
       }, renderAndRun);
     }
-  }).version("1.45.4 (UR-Nexus)", "-v, --version", "Output the version number");
+  }).version("1.45.5 (UR-Nexus)", "-v, --version", "Output the version number");
   program2.option("-w, --worktree [name]", "Create a new git worktree for this session (optionally specify a name)");
   program2.option("--tmux", "Create a tmux session for the worktree (requires --worktree). Uses iTerm2 native panes when available; use --tmux=classic for traditional tmux.");
   if (canUserConfigureAdvisor()) {
@@ -706042,7 +706102,7 @@ if (false) {}
 async function main2() {
   const args = process.argv.slice(2);
   if (args.length === 1 && (args[0] === "--version" || args[0] === "-v" || args[0] === "-V")) {
-    console.log(`${"1.45.4"} (UR-Nexus)`);
+    console.log(`${"1.45.5"} (UR-Nexus)`);
     return;
   }
   if (args[0] === "a2a" && args[1] === "serve" && !args.includes("--help") && !args.includes("-h")) {

package/docs/AGENT_FEATURES.md

@@ -9,6 +9,15 @@ reproducible autonomous software engineering agent: every substantial task can
 be driven as `spec -> plan -> patch -> test -> report -> rollback`, with the
 spec as the durable source of truth and command evidence as the success gate.
 
+## v1.45.5 Additions
+
+- Ollama Cloud requests have bounded response-header and streaming phases and
+  do not amplify a deliberate stream timeout through fallback retries. Local
+  model timing and explicit timeout overrides are unchanged.
+- The verifier now recognizes terminal promises such as "Let me create it
+  now" or "I will run the tests now" and requires the corresponding successful
+  mutation or Bash call before the turn may complete.
+
 ## Commands
 
 ```sh

package/docs/USAGE.md

@@ -100,6 +100,12 @@ use `ollama.host` in settings if you want plain `ur` to default to a LAN host.
 Models exposed by the chosen Ollama app are valid, including local models and
 Ollama Cloud-backed models.
 
+Ollama Cloud models use a 120-second default bound for both response-header
+waiting and stream consumption. A deliberate stream deadline is returned
+directly instead of being replayed through the non-streaming fallback. Local
+Ollama models keep the five-minute default. Set `API_TIMEOUT_MS` to explicitly
+override either default for the current process.
+
 UR-Nexus also has explicit provider commands for legal access paths:
 
 ```sh

package/docs/VALIDATION.md

@@ -19,7 +19,7 @@ You need:
 
 ```sh
 ur --version
-# expected for this release: "1.45.4 (UR-Nexus)"
+# expected for this release: "1.45.5 (UR-Nexus)"
 ```
 
 ## 0.1 First-workspace model selection (1.45.4)
@@ -40,6 +40,20 @@ ur -p "say hi"
 Expected: `No model has been selected for this workspace`. Supplying
 `--model <model>`, `OLLAMA_MODEL`, or `UR_MODEL` bypasses the gate for that run.
 
+### 0.1.1 Ollama Cloud latency containment (1.45.5)
+
+Run the deterministic regression coverage:
+
+```sh
+bun test test/ollamaTimeout.test.ts
+```
+
+Expected: the cloud-model timeout, override-precedence, stream-deadline, and
+fallback-suppression cases pass. A model ending in `:cloud` defaults to 120
+seconds for response headers and streaming; a local model retains 300 seconds.
+`API_TIMEOUT_MS` wins over both. When a cloud stream reaches its deliberate
+deadline, the request fails once instead of starting a non-streaming replay.
+
 ## 0.2 Permission safety and context pack (1.19.0)
 
 In a project checkout:
@@ -186,6 +200,11 @@ Expected:
   UR_VERIFIER_MODE=strict ur  # default, false claim rejected
   ```
 
+Also prompt the model to end with `Let me create index.html now.` while
+forbidding tool calls. Strict mode must inject a correction and continue rather
+than accepting the promise as completion. Conditional text such as `If you
+approve, I'll create it` must not trigger this gate.
+
 ## 3. L1 loop detector fires
 
 ```text

package/docs/providers.md

@@ -431,6 +431,13 @@ Local/server providers use their normal endpoints:
 - llama.cpp server mode: `http://localhost:8080/v1`
 - vLLM server mode: `http://localhost:8000/v1`
 
+Ollama models whose names end in `:cloud` use a 120-second default for the
+response-header phase and a 120-second total stream deadline. UR does not
+automatically replay a cloud request after that stream deadline, preventing a
+bounded failure from expanding into the shared non-streaming fallback and
+retry chain. Local Ollama models retain the five-minute default. A positive
+`API_TIMEOUT_MS` or explicit request timeout overrides these defaults.
+
 ## Optional Live Provider Smoke
 
 `bun run provider:smoke` runs optional live checks only for providers with the

package/documentation/index.html

@@ -44,7 +44,7 @@
     <main id="content" class="content">
       <header class="topbar">
         <div>
-          <p class="eyebrow">Version 1.45.4</p>
+          <p class="eyebrow">Version 1.45.5</p>
           <h1>UR-Nexus Documentation</h1>
           <p class="lead">A practical, tutorial-style reference for installing, configuring, automating, extending, and operating UR-Nexus.</p>
         </div>
@@ -163,9 +163,14 @@ ur config set provider.fallback ollama</code></pre>
           </article>
           <article>
             <h3>Status bar and updates</h3>
-            <pre><code>Ollama | llama3 | ask | main | update 1.45.4 available</code></pre>
+            <pre><code>Ollama | llama3 | ask | main | update 1.45.5 available</code></pre>
             <p>The interactive status bar shows only important runtime state: provider, model, mode, branch, active tasks, checks status when known, and update availability. It is hidden in CI, dumb terminals, and print mode.</p>
           </article>
+          <article>
+            <h3>Ollama Cloud latency bounds</h3>
+            <pre><code>API_TIMEOUT_MS=120000 ur</code></pre>
+            <p>Cloud-tagged Ollama models default to bounded response-header and streaming phases. A stream deadline is surfaced directly instead of starting a non-streaming retry chain. Local models retain their longer default, and <code>API_TIMEOUT_MS</code> remains an explicit override.</p>
+          </article>
         </div>
       </section>
 
@@ -481,7 +486,7 @@ ur context-pack compress</code></pre>
 UR_VERIFIER_MODE=loose
 UR_VERIFIER_MODE=off
 UR_VERIFIER_AUTO_SUBAGENT=1</code></pre>
-            <p>L1 gates catch false done claims and loops. <code>/verify</code> manually runs deeper verification.</p>
+            <p>L1 gates catch false done claims, immediate-action promises that end without their tool call, and loops. <code>/verify</code> manually runs deeper verification.</p>
             <p>Interactive users can set <code>"verifier": { "askBeforeGates": true }</code> in <code>.ur/settings.json</code> (or via <code>ur config set verifier.askBeforeGates true</code>) to have UR ask before running project test/typecheck/lint gates after a task, instead of running them automatically. Default is <code>false</code> (auto-run gates).</p>
           </article>
           <article>

package/extensions/jetbrains-ur/build.gradle.kts

@@ -5,7 +5,7 @@ plugins {
 }
 
 group = "dev.urnexus"
-version = "1.45.4"
+version = "1.45.5"
 
 repositories {
     mavenCentral()

package/extensions/vscode-ur-inline-diffs/package.json

@@ -2,7 +2,7 @@
   "name": "ur-inline-diffs",
   "displayName": "UR Inline Diffs",
   "description": "Review, apply, and reject UR inline diff bundles from .ur/ide/diffs inside VS Code.",
-  "version": "1.45.4",
+  "version": "1.45.5",
   "publisher": "ur-nexus",
   "engines": {
     "vscode": "^1.92.0"

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ur-agent",
-  "version": "1.45.4",
+  "version": "1.45.5",
   "description": "UR-Nexus — autonomous engineering workflow engine (plan, execute, test, verify, document, benchmark, reproduce)",
   "type": "module",
   "packageManager": "bun@1.3.14",