@cartanova/qgrid-cli 2.10.0 → 2.10.2

package/bundle/web-dist/server/entry-server.generated.js

@@ -1,6 +1,6 @@
-import { A as hashQueryKeyByOptions, B as __exportAll, C as Mutation, D as notifyManager, E as onlineManager, F as resolveStaleTime, H as __toESM, I as skipToken, L as focusManager, M as matchQuery, N as noop$1, O as functionalUpdate, P as partialMatchKey, R as Subscribable, S as require_react, T as Query, V as __require, d as setLocale, f as _enum, h as string$2, j as matchMutation, k as hashKey, l as SD, m as object, n as QgridService, o as TokenService, p as number$1, s as fetch$1, u as SUPPORTED_LOCALES, w as infiniteQueryBehavior, x as require_jsx_runtime, y as QueryClientProvider, z as __commonJSMin } from "./assets/services.generated-BsKOUSXl.js";
-import { A as decodePath, C as findRouteMatch, D as createLRUCache, E as processRouteTree, F as last, I as replaceEqualDeep, M as findLast, N as functionalUpdate$1, O as invariant, P as isPromise, S as findFlatMatch, T as processRouteMasks, _ as interpolatePath, a as Link$1, b as trimPath, c as usePrevious$1, d as useRouter, f as getRouterContext, g as cleanPath, h as isNotFound, i as createRootRouteWithContext, j as deepEqual$2, k as createControlledPromise, l as matchContext, m as rootRouteId, n as createFileRoute, p as warning, r as createLazyFileRoute, s as useLayoutEffect$3, t as lazyRouteComponent, u as useRouterState, v as joinPaths, w as findSingleMatch, x as trimPathRight, y as resolvePath } from "./assets/lazyRouteComponent-DyizoQKi.js";
-import { n as require_shim, t as require_react_dom } from "./assets/react-dom-PfWvArTX.js";
+import { A as require_jsx_runtime, B as matchMutation, C as union, F as onlineManager, G as skipToken, H as noop$1, I as notifyManager, J as __commonJSMin, K as focusManager, L as functionalUpdate, M as Mutation, N as infiniteQueryBehavior, O as QueryClientProvider, P as Query, R as hashKey, S as string$2, U as partialMatchKey, V as matchQuery, W as resolveStaleTime, X as __require, Y as __exportAll, Z as __toESM, _ as custom, b as number$1, c as fetch$1, d as SD, f as SUPPORTED_LOCALES, g as boolean$1, h as array, j as require_react, m as _enum, n as QgridService, o as TokenService, p as setLocale, q as Subscribable, s as SonamuQueryMode, u as zArrayable, v as date, w as unknown, x as object, y as int, z as hashQueryKeyByOptions } from "./assets/services.generated-BM4X3Ecj.js";
+import { A as decodePath, C as findRouteMatch, D as createLRUCache, E as processRouteTree, F as last, I as replaceEqualDeep, M as findLast, N as functionalUpdate$1, O as invariant, P as isPromise, S as findFlatMatch, T as processRouteMasks, _ as interpolatePath, a as Link$1, b as trimPath, c as usePrevious$1, d as useRouter, f as getRouterContext, g as cleanPath, h as isNotFound, i as createRootRouteWithContext, j as deepEqual$2, k as createControlledPromise, l as matchContext, m as rootRouteId, n as createFileRoute, p as warning, r as createLazyFileRoute, s as useLayoutEffect$3, t as lazyRouteComponent, u as useRouterState, v as joinPaths, w as findSingleMatch, x as trimPathRight, y as resolvePath } from "./assets/lazyRouteComponent-Bj1cPq6x.js";
+import { n as require_shim, t as require_react_dom } from "./assets/react-dom-BjR1Iaz8.js";
 import minpath from "node:path";
 import minproc from "node:process";
 import { fileURLToPath as urlToPath } from "node:url";
@@ -46269,12 +46269,12 @@ var MODEL_PRESET_GROUPS = [{
 		"anthropic/claude-opus-5",
 		"anthropic/claude-sonnet-5-5",
 		"anthropic/claude-sonnet-5",
+		"anthropic/claude-haiku-5-5",
 		"anthropic/claude-opus-4-8",
 		"anthropic/claude-opus-4-7",
 		"anthropic/claude-opus-4-6",
 		"anthropic/claude-opus-4-5",
 		"anthropic/claude-sonnet-4-6",
-		"anthropic/claude-sonnet-4-5",
 		"anthropic/claude-haiku-4-5"
 	]
 }];
@@ -46393,7 +46393,7 @@ function ChatWidget() {
 					tokenName: data.tokenName,
 					durationMs: data.durationMs,
 					costUsd: data.costUsd,
-					requestLogId: data.runContext?.requestLogId,
+					requestLogId: data.requestLogId,
 					fallback: (data.modelFallbacks?.length ?? 0) > 0
 				}
 			}));
@@ -46492,7 +46492,7 @@ function ChatWidget() {
 						tokenName: result.tokenName,
 						durationMs: result.durationMs,
 						costUsd: result.costUsd,
-						requestLogId: result.runContext?.requestLogId,
+						requestLogId: result.requestLogId,
 						fallback: (result.modelFallbacks?.length ?? 0) > 0
 					}
 				}));
@@ -46874,7 +46874,7 @@ function ChatWidget() {
 }
 //#endregion
 //#region src/components/Sidebar.tsx
-var qgridVersion = "2.10.0";
+var qgridVersion = "2.10.2";
 function Sidebar({ className }) {
 	const pathname = useRouterState({ select: (s) => s.location.pathname });
 	return /* @__PURE__ */ (0, import_jsx_runtime.jsxs)("aside", {
@@ -68697,15 +68697,9 @@ var Lo = import_react.forwardRef(({ className: t, children: n, variant: r, size:
 Lo.displayName = Item2.displayName;
 //#endregion
 //#region src/contexts/sonamu-provider.tsx
-var uploader = async (files) => {
-	if (files.length === 0) return [];
-	console.log("File upload not implemented yet");
-	return [];
-};
 var sd = (key) => SD(key);
 function SonamuProvider({ children }) {
 	return /* @__PURE__ */ (0, import_jsx_runtime.jsx)(qt, {
-		uploader,
 		SD: sd,
 		children
 	});
@@ -68758,15 +68752,15 @@ function RootComponent() {
 }
 //#endregion
 //#region src/routes/tokens.tsx
-var $$splitComponentImporter$5 = () => import("./assets/tokens-DAUmTrgO.js");
+var $$splitComponentImporter$5 = () => import("./assets/tokens-DiTUwUwr.js");
 var Route$5 = createFileRoute("/tokens")({ component: lazyRouteComponent($$splitComponentImporter$5, "component") });
 //#endregion
 //#region src/routes/settings.tsx
-var $$splitComponentImporter$4 = () => import("./assets/settings-DPcWGvy-.js");
+var $$splitComponentImporter$4 = () => import("./assets/settings-BFVuldch.js");
 var Route$4 = createFileRoute("/settings")({ component: lazyRouteComponent($$splitComponentImporter$4, "component") });
 //#endregion
 //#region src/routes/monit.tsx
-var $$splitComponentImporter$3 = () => import("./assets/monit-COH4MckI.js");
+var $$splitComponentImporter$3 = () => import("./assets/monit-D9W4ioUi.js");
 var monitSearchSchema = object({ level: _enum([
 	"all",
 	"warn",
@@ -68777,14 +68771,379 @@ var Route$3 = createFileRoute("/monit")({
 	component: lazyRouteComponent($$splitComponentImporter$3, "component")
 });
 //#endregion
+//#region src/services/sonamu.generated.ts
+/**
+* @generated
+* API에서 동기화된 파일입니다. 직접 수정하지 마세요.
+*/
+var HistoryItems = array(object({
+	type: string$2(),
+	role: string$2().optional(),
+	content: unknown().optional(),
+	name: string$2().optional(),
+	arguments: string$2().optional(),
+	call_id: string$2().optional(),
+	output: string$2().optional()
+}));
+var TokenCredentials = union([object({
+	accessToken: string$2(),
+	refreshToken: string$2(),
+	expiresAt: number$1(),
+	accountUuid: string$2()
+}), object({
+	accessToken: string$2(),
+	refreshToken: string$2(),
+	idToken: string$2().optional(),
+	accessTokenExpiresAt: number$1(),
+	idTokenExpiresAt: number$1().optional(),
+	accountId: string$2(),
+	planType: string$2().optional()
+})]);
+var ToolDefinitions = array(object({
+	name: string$2(),
+	description: string$2().optional(),
+	inputSchema: unknown()
+}));
+var RequestLogOrderBy = _enum([
+	"id-desc",
+	"id-asc",
+	"created_at-desc",
+	"created_at-asc",
+	"ttft_ms-desc",
+	"ttft_ms-asc",
+	"duration_ms-desc",
+	"duration_ms-asc",
+	"input_tokens-desc",
+	"input_tokens-asc",
+	"output_tokens-desc",
+	"output_tokens-asc",
+	"cache_read_tokens-desc",
+	"cache_read_tokens-asc",
+	"cache_creation_tokens-desc",
+	"cache_creation_tokens-asc",
+	"cost_usd-desc",
+	"cost_usd-asc"
+]).describe("RequestLogOrderBy");
+var RequestLogSearchField = _enum([
+	"id",
+	"token_name",
+	"user_prompt"
+]).describe("RequestLogSearchField");
+var RequestLogStatus = _enum([
+	"running",
+	"succeeded",
+	"error",
+	"aborted"
+]).describe("RequestLogStatus");
+var RequestLogStepType = _enum(["generate", "tool_call"]).describe("RequestLogStepType");
+var RequestLogStepOrderBy = _enum(["id-asc"]).describe("RequestLogStepOrderBy");
+var RequestLogStepSearchField = _enum(["id"]).describe("RequestLogStepSearchField");
+var SettingOrderBy = _enum(["key-asc"]).describe("SettingOrderBy");
+var SettingSearchField = _enum(["key"]).describe("SettingSearchField");
+var TokenOrderBy = _enum(["id-desc", "ord-asc"]).describe("TokenOrderBy");
+var TokenSearchField = _enum(["id", "name"]).describe("TokenSearchField");
+object({
+	id: int(),
+	created_at: date(),
+	token_name: string$2().max(100).nullable(),
+	project_name: string$2().max(50).nullable(),
+	model_name: string$2().max(255).nullable(),
+	requested_model_name: string$2().max(255).nullable(),
+	fallback_count: int().nullable(),
+	user_prompt: string$2().nullable(),
+	system_prompt: string$2().nullable(),
+	response: string$2(),
+	input_tokens: int(),
+	output_tokens: int(),
+	cache_read_tokens: int(),
+	cache_creation_tokens: int(),
+	cache_creation_5m_tokens: int().nullable(),
+	cache_creation_1h_tokens: int().nullable(),
+	duration_ms: int(),
+	ttft_ms: int(),
+	cost_usd: int().nullable(),
+	cost_source: string$2().max(20).nullable(),
+	image_cost_usd: int().nullable(),
+	image_cost_method: string$2().max(100).nullable(),
+	effort: string$2().max(10).nullable(),
+	history: HistoryItems.nullable(),
+	tools: ToolDefinitions.nullable(),
+	json_schema: string$2().nullable(),
+	status: RequestLogStatus,
+	error_message: string$2().nullable(),
+	response_json_ok: boolean$1().nullable(),
+	tool_call_count: int(),
+	is_image_generation: boolean$1(),
+	is_structured: boolean$1()
+});
+object({
+	id: int(),
+	created_at: date(),
+	request_log_id: int(),
+	step_index: int(),
+	type: RequestLogStepType,
+	model_name: string$2().max(255).nullable(),
+	requested_model_name: string$2().max(255).nullable(),
+	fallback_count: int().nullable(),
+	input_tokens: int().nullable(),
+	output_tokens: int().nullable(),
+	cache_read_tokens: int().nullable(),
+	cache_creation_tokens: int().nullable(),
+	cache_creation_5m_tokens: int().nullable(),
+	cache_creation_1h_tokens: int().nullable(),
+	cost_usd: int().nullable(),
+	cost_source: string$2().max(20).nullable(),
+	duration_ms: int().nullable(),
+	ttft_ms: int().nullable(),
+	finish_reason: string$2().max(20).nullable(),
+	reasoning_text: string$2().nullable(),
+	reasoning_tokens: int().nullable(),
+	tool_call_index: int().nullable(),
+	tool_call_id: string$2().max(100).nullable(),
+	tool_name: string$2().max(100).nullable(),
+	tool_args: string$2().nullable(),
+	tool_result: string$2().nullable(),
+	tool_duration_ms: int().nullable(),
+	error: string$2().nullable()
+});
+object({
+	id: int(),
+	created_at: date(),
+	key: string$2().max(100),
+	value: string$2(),
+	updated_at: date().nullable()
+});
+object({
+	id: int(),
+	created_at: date(),
+	provider: string$2().max(20),
+	credentials: TokenCredentials,
+	name: string$2(),
+	active: boolean$1(),
+	reauth_required: boolean$1(),
+	ord: int(),
+	quota_threshold: int().nullable(),
+	weight: int(),
+	keepalive_enabled: boolean$1()
+});
+var RequestLogBaseListParams = object({
+	num: number$1().int().nonnegative(),
+	page: number$1().int().min(1),
+	search: RequestLogSearchField,
+	keyword: string$2(),
+	orderBy: RequestLogOrderBy,
+	queryMode: SonamuQueryMode,
+	id: zArrayable(number$1().int().positive()),
+	sonamuFilter: custom(),
+	token_name: string$2().max(100).nullable(),
+	project_name: string$2().max(50).nullable(),
+	model_name: string$2().max(255).nullable()
+}).partial();
+object({
+	num: number$1().int().nonnegative(),
+	page: number$1().int().min(1),
+	search: RequestLogStepSearchField,
+	keyword: string$2(),
+	orderBy: RequestLogStepOrderBy,
+	queryMode: SonamuQueryMode,
+	id: zArrayable(number$1().int().positive()),
+	sonamuFilter: custom(),
+	request_log_id: int()
+}).partial();
+object({
+	num: number$1().int().nonnegative(),
+	page: number$1().int().min(1),
+	search: SettingSearchField,
+	keyword: string$2(),
+	orderBy: SettingOrderBy,
+	queryMode: SonamuQueryMode,
+	id: zArrayable(number$1().int().positive()),
+	sonamuFilter: custom(),
+	key: string$2().max(100)
+}).partial();
+object({
+	num: number$1().int().nonnegative(),
+	page: number$1().int().min(1),
+	search: TokenSearchField,
+	keyword: string$2(),
+	orderBy: TokenOrderBy,
+	queryMode: SonamuQueryMode,
+	id: zArrayable(number$1().int().positive()),
+	sonamuFilter: custom()
+}).partial();
+object({
+	id: int(),
+	created_at: date(),
+	token_name: string$2().max(100).nullable(),
+	project_name: string$2().max(50).nullable(),
+	model_name: string$2().max(255).nullable(),
+	requested_model_name: string$2().max(255).nullable(),
+	fallback_count: int().nullable(),
+	user_prompt: string$2().nullable(),
+	system_prompt: string$2().nullable(),
+	response: string$2(),
+	input_tokens: int(),
+	output_tokens: int(),
+	cache_read_tokens: int(),
+	cache_creation_tokens: int(),
+	cache_creation_5m_tokens: int().nullable(),
+	cache_creation_1h_tokens: int().nullable(),
+	duration_ms: int(),
+	ttft_ms: int(),
+	cost_usd: int().nullable(),
+	cost_source: string$2().max(20).nullable(),
+	image_cost_usd: int().nullable(),
+	image_cost_method: string$2().max(100).nullable(),
+	effort: string$2().max(10).nullable(),
+	history: HistoryItems.nullable(),
+	tools: ToolDefinitions.nullable(),
+	status: RequestLogStatus,
+	response_json_ok: boolean$1().nullable(),
+	error_message: string$2().nullable(),
+	tool_call_count: int(),
+	is_image_generation: boolean$1(),
+	is_structured: boolean$1()
+});
+object({
+	id: int(),
+	created_at: date(),
+	token_name: string$2().max(100).nullable(),
+	project_name: string$2().max(50).nullable(),
+	model_name: string$2().max(255).nullable(),
+	requested_model_name: string$2().max(255).nullable(),
+	fallback_count: int().nullable(),
+	input_tokens: int(),
+	output_tokens: int(),
+	cache_read_tokens: int(),
+	cache_creation_tokens: int(),
+	cache_creation_5m_tokens: int().nullable(),
+	cache_creation_1h_tokens: int().nullable(),
+	duration_ms: int(),
+	ttft_ms: int(),
+	cost_usd: int().nullable(),
+	cost_source: string$2().max(20).nullable(),
+	image_cost_usd: int().nullable(),
+	image_cost_method: string$2().max(100).nullable(),
+	effort: string$2().max(10).nullable(),
+	status: RequestLogStatus,
+	response_json_ok: boolean$1().nullable(),
+	tool_call_count: int(),
+	is_image_generation: boolean$1(),
+	is_structured: boolean$1()
+});
+object({
+	id: int(),
+	token_name: string$2().max(100).nullable(),
+	model_name: string$2().max(255).nullable(),
+	input_tokens: int(),
+	output_tokens: int(),
+	cache_read_tokens: int(),
+	cache_creation_tokens: int(),
+	cache_creation_5m_tokens: int().nullable(),
+	cache_creation_1h_tokens: int().nullable(),
+	cost_usd: int().nullable(),
+	cost_source: string$2().max(20).nullable()
+});
+_enum([
+	"A",
+	"P",
+	"C"
+]);
+object({
+	id: int(),
+	created_at: date(),
+	step_index: int(),
+	type: RequestLogStepType,
+	model_name: string$2().max(255).nullable(),
+	requested_model_name: string$2().max(255).nullable(),
+	fallback_count: int().nullable(),
+	input_tokens: int().nullable(),
+	output_tokens: int().nullable(),
+	cache_read_tokens: int().nullable(),
+	cache_creation_tokens: int().nullable(),
+	cache_creation_5m_tokens: int().nullable(),
+	cache_creation_1h_tokens: int().nullable(),
+	cost_usd: int().nullable(),
+	cost_source: string$2().max(20).nullable(),
+	duration_ms: int().nullable(),
+	ttft_ms: int().nullable(),
+	finish_reason: string$2().max(20).nullable(),
+	reasoning_text: string$2().nullable(),
+	reasoning_tokens: int().nullable(),
+	tool_call_index: int().nullable(),
+	tool_call_id: string$2().max(100).nullable(),
+	tool_name: string$2().max(100).nullable(),
+	tool_args: string$2().nullable(),
+	tool_result: string$2().nullable(),
+	tool_duration_ms: int().nullable(),
+	error: string$2().nullable()
+});
+object({
+	id: int(),
+	created_at: date(),
+	step_index: int(),
+	type: RequestLogStepType,
+	model_name: string$2().max(255).nullable(),
+	requested_model_name: string$2().max(255).nullable(),
+	fallback_count: int().nullable(),
+	input_tokens: int().nullable(),
+	output_tokens: int().nullable(),
+	cache_read_tokens: int().nullable(),
+	cache_creation_tokens: int().nullable(),
+	cache_creation_5m_tokens: int().nullable(),
+	cache_creation_1h_tokens: int().nullable(),
+	cost_usd: int().nullable(),
+	cost_source: string$2().max(20).nullable(),
+	duration_ms: int().nullable(),
+	ttft_ms: int().nullable(),
+	finish_reason: string$2().max(20).nullable(),
+	reasoning_tokens: int().nullable(),
+	tool_call_index: int().nullable(),
+	tool_call_id: string$2().max(100).nullable(),
+	tool_name: string$2().max(100).nullable(),
+	tool_duration_ms: int().nullable(),
+	error: string$2().nullable()
+});
+object({
+	id: int(),
+	tool_args: string$2().nullable()
+});
+_enum([
+	"A",
+	"T",
+	"I"
+]);
+object({
+	id: int(),
+	created_at: date(),
+	key: string$2().max(100),
+	value: string$2(),
+	updated_at: date().nullable()
+});
+_enum(["A"]);
+object({
+	id: int(),
+	created_at: date(),
+	provider: string$2().max(20),
+	credentials: TokenCredentials,
+	name: string$2(),
+	active: boolean$1(),
+	reauth_required: boolean$1(),
+	ord: int(),
+	quota_threshold: int().nullable(),
+	weight: int(),
+	keepalive_enabled: boolean$1()
+});
+_enum(["A"]);
+//#endregion
 //#region src/routes/logs.tsx
-var $$splitComponentImporter$2 = () => import("./assets/logs-DO8zBdBR.js");
+var $$splitComponentImporter$2 = () => import("./assets/logs-BSk9lOHe.js");
 var logsSearchSchema = object({
 	token: string$2().optional(),
 	project: string$2().optional(),
 	model: string$2().optional(),
-	sort: string$2().optional(),
-	page: number$1().optional().default(1)
+	sort: RequestLogOrderBy.optional(),
+	page: RequestLogBaseListParams.shape.page.default(1)
 });
 var Route$2 = createFileRoute("/logs")({
 	validateSearch: logsSearchSchema,
@@ -68792,11 +69151,11 @@ var Route$2 = createFileRoute("/logs")({
 });
 //#endregion
 //#region src/routes/index.tsx
-var $$splitComponentImporter$1 = () => import("./assets/routes-Btja-741.js");
+var $$splitComponentImporter$1 = () => import("./assets/routes-C7lB2Q85.js");
 var Route$1 = createFileRoute("/")({ component: lazyRouteComponent($$splitComponentImporter$1, "component") });
 //#endregion
 //#region src/routes/requests/show.tsx
-var $$splitComponentImporter = () => import("./assets/show-CMXbb3TJ.js");
+var $$splitComponentImporter = () => import("./assets/show-Cswh_tqs.js");
 var showSearchSchema = object({ id: number$1() });
 var Route = createFileRoute("/requests/show")({
 	validateSearch: showSearchSchema,
@@ -68871,4 +69230,4 @@ async function render(url, preloadedData = []) {
 	};
 }
 //#endregion
-export { clsx as a, remarkGfm as c, Q as i, Markdown as l, Route$2 as n, lucideX as o, Route$3 as r, render, lucideRotateCcw as s, Route as t, lucideKeyRound as u };
+export { Route$3 as a, lucideX as c, Markdown as d, lucideKeyRound as f, RequestLogOrderBy as i, lucideRotateCcw as l, Route$2 as n, Q as o, RequestLogBaseListParams as r, render, clsx as s, Route as t, remarkGfm as u };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@cartanova/qgrid-cli",
-  "version": "2.10.0",
+  "version": "2.10.2",
   "repository": {
     "type": "git",
     "url": "https://github.com/cartanova-ai/qgrid"

package/skills/qgrid/references/ai-sdk-provider-contract.md

@@ -68,7 +68,7 @@ Do not write `providerOptions: { ... } satisfies QgridProviderOptions`: the expo
 - `reasoningSummary`: OpenAI/Codex route only.
 - `serviceTier`: OpenAI/Codex route only.
 - `timeoutMs`: Anthropic server-side Claude Code process timeout in milliseconds. It must be a positive integer no greater than 30 minutes and defaults to 240 seconds. For non-stream `generateText`, the SDK derives request-scoped Undici `headersTimeout` and `bodyTimeout` values as `timeoutMs + 60_000` without changing the process-global dispatcher.
-- `fallbackModels`: reserved for future qgrid server-side fallback routing. It is not the Fable 5 safety-refusal fallback, which is owned by Claude Code upstream.
+- `fallbackModels`: deprecated, ignored compatibility field. The provider emits a standard unsupported-option warning when supplied; remove it from callers. It is not Claude Code's upstream Fable safety-refusal fallback.
 - `imageGeneration`: OpenAI/Codex non-stream only. Enables Codex's built-in `image_generation` tool for that request.
 - `imageGenerationOptions`: optional `quality: "low" | "medium" | "high"`, `size: "1024x1024" | "1024x1536" | "1536x1024"`, and `background: "auto" | "opaque" | "transparent"`. Transparent requests use Codex standalone Images with `gpt-image-2`; other requests use the hosted Responses tool. Transparent mode rejects tools/structured output, supports up to five references, validates real transparent and visible PNG pixels, and never invokes background-removal fallback. Returned size/quality may differ from requested settings; read `result.providerMetadata.qgrid.imageGeneration` or per-file `result.content` metadata. `result.files[]` contains bytes/mediaType only. Logs separate `requestedOptions` from `observedGeneration`; image usage is separate from the zero text-driver usage on this route.
 
@@ -199,6 +199,8 @@ Reference images for image generation use normal AI SDK multimodal message parts
 
 ## Response mapping
 
+Cancelling the returned provider stream aborts its HTTP request and releases the SSE reader; caller abort signals propagate through the same request-local cancellation path. Normal completion and error events also release the reader. OpenAI affinity coordinates are swept for expiration on insertion and bounded to the most recently issued 1,000 entries; eviction keeps full-history generation functional.
+
 Since 2.10.0, `generateText` exposes the parent request-log ID at `result.providerMetadata?.qgrid?.requestLogId`. For `streamText`, consume the stream first, then read `(await result.providerMetadata)?.qgrid?.requestLogId`. The raw query response and stream `done` event carry optional top-level `requestLogId`.
 
 Update both server and SDK to use this metadata. The field is omitted when logging is disabled or an older server does not return it. Tool-run steps share the parent ID; `runContext.requestLogId` remains separate continuation state. This does not expose external-provider telemetry logger IDs or add IDs to stream delta/error events.
@@ -226,7 +228,7 @@ For a successful Fable refusal fallback (Fable 5 or 5.1):
 - `providerMetadata.qgrid.modelFallbacks` preserves the refusal route and optional category/explanation.
 - `providerMetadata.qgrid.costSource` reports whether cost came from Claude Code or qgrid's pricing table. Prefer the provider-reported combined cost for this path.
 
-Do not expose the requested Fable model as `response.modelId` after Opus served the answer, and do not map this upstream safety behavior onto the reserved `providerOptions.qgrid.fallbackModels` option.
+Do not expose the requested Fable model as `response.modelId` after Opus served the answer, and do not map this upstream safety behavior onto the deprecated `providerOptions.qgrid.fallbackModels` option.
 
 Usage maps qgrid standard usage into AI SDK V3 usage:
 

package/skills/qgrid/references/anthropic-claude-code-runtime.md

@@ -77,7 +77,7 @@ claude -p
   --model <canonical-model-or-1m-suffix>
   --system-prompt <text>                  # small system prompt
   --system-prompt-file <path>             # large system prompt
-  --thinking disabled                       # omitted for Fable 5/5.1, Opus 5, Opus 5.5 and Sonnet 5.5
+  --thinking disabled                       # omitted for Fable 5/5.1, Opus 5, Opus 5.5, Sonnet 5.5 and Haiku 5.5
   --effort <effort-or-low>
   --disable-slash-commands
   --session-id <uuid>
@@ -97,7 +97,7 @@ Important details:
 - Large system prompts over 64 KiB are written to a temporary file to avoid argv `E2BIG` — this
   same branch absorbs large injected schema contracts, so the old 64 KiB schema argv limit no
   longer applies (only the global 512 KiB caller-schema complexity limit remains).
-- `--thinking disabled`, `MAX_THINKING_TOKENS=0`, and adaptive thinking env suppression keep thinking off for existing models. Fable 5, Fable 5.1, Opus 5.5 and Sonnet 5.5 require always-on adaptive thinking (Fable: CLI catalog `rejects_disabled_thinking`; Opus 5.5: official spec, thinking cannot be turned off; Sonnet 5.5: API `{type: "disabled"}` returns 400 and the CLI catalog marks `rejects_disabled_thinking`). Opus 5 defaults to adaptive thinking and rejects disabled thinking at `xhigh`/`max` effort. qgrid omits all three suppressors for these models (`usesAdaptiveThinking`) and uses `effort` to control depth.
+- `--thinking disabled`, `MAX_THINKING_TOKENS=0`, and adaptive thinking env suppression keep thinking off for existing models. Fable 5, Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5 require always-on adaptive thinking (Fable: CLI catalog `rejects_disabled_thinking`; Opus 5.5: official spec, thinking cannot be turned off; Sonnet 5.5: API `{type: "disabled"}` returns 400 and the CLI catalog marks `rejects_disabled_thinking`; Haiku 5.5 (released 2026-10-07): the API accepts disabled thinking only at `high` effort or below and returns 400 at `xhigh`/`max`, and the Claude Code 2.1.293 catalog marks `rejects_disabled_thinking`). Opus 5 defaults to adaptive thinking and rejects disabled thinking at `xhigh`/`max` effort. qgrid omits all three suppressors for these models (`usesAdaptiveThinking`) and uses `effort` to control depth.
 
 ## Spawn env
 
@@ -260,6 +260,7 @@ qgrid's exact 1M support set currently includes:
 - `claude-opus-5-5`
 - `claude-sonnet-5-5`
 - `claude-sonnet-5`
+- `claude-haiku-5-5`
 - `claude-sonnet-4-6`
 - `claude-opus-4-6`
 - `claude-opus-4-8`

package/skills/qgrid/references/request-log-run-lifecycle.md

@@ -63,6 +63,8 @@ Provider-qualified ids widen `model_name` and `requested_model_name` from 50 to
 
 ## Usage fields
 
+The dashboard total includes driver `cost_usd` and `image_cost_usd` once each. Both the stored sum and legacy driver-price adjustment apply the same normalized Sonamu filters as the list. The `/api/qgrid/stats` `requests` field is an in-memory count of completed provider generations since process start, separated by provider and token name; tool follow-up generations count separately. It is not a persisted lifetime count or a count of failed attempts.
+
 Stored request log usage uses qgrid-standard semantics:
 
 - `input_tokens`: total input, including cache read/write.

package/skills/qgrid/references/sonamu-api-web-flow.md

@@ -60,6 +60,8 @@ Image-generation request logs add `is_image_generation`, `image_cost_usd`, and `
 
 In 2.9.8, list and detail model labels show requested model to image model for generated-image results, using an image serving model or the model recorded in `image_cost_method`. The cost fallback is an estimate, not an upstream version guarantee. Preserve the actual driver in stored `model_name` for hosted/auto turns. Do not hardcode a display version or render an image-generation/assumption badge. Auto text-only replies are not marked `is_image_generation`.
 
+The list header ends its filter row with a reload button (2.10.2+) that calls the generated hooks' `refresh()` for the list and the total-cost query with the current filters; it does not refetch the token/project/model filter options, and TanStack Query's `isRefreshing` disables it while a reload is in flight.
+
 Structured output visibility: the list has an `Fmt` column (`json` badge from `is_structured` vs `text`) and a `broken` badge driven by `response_json_ok = false`; there is no broken-JSON list filter. At review time, all 114,647 succeeded structured rows had `response_json_ok = true` and zero false rows existed; the unindexed zero-result lookup scanned 1,286,016 rows and took about 3 seconds. Before reinstating the filter, validate the failure-signal semantics and index the predicate. The detail page shows a dark Response Type panel (zod expression by default, `type` declaration via toggle) rendered server-side from the stored `json_schema`. Plain-text rows render no panel. Rows written by pre-2.7.2 servers show `text`.
 
 Tool contract visibility: the detail page renders a Tools panel between Response Type and Steps for requests whose `tools` is non-empty. It starts collapsed and its header shows the tool count plus a responsive, truncated name list built from the detail subset — collapsed views issue no extra query. Expanding fires `toolsView` once and caches it for that request id, since a stored tool contract never changes. Each card shows the name, the full description, the parameter count, and a compact display-only zod shape; enum values past the sixth collapse to `/* +N */`, with the full shape reachable through hover and keyboard focus. All JSON Schema conversion happens server-side — the web renders completed strings and holds no schema logic.
@@ -72,7 +74,7 @@ Clicking an attached or generated image opens a preview with extension, decoded
 
 Token credential previews must tolerate missing `accessToken` rather than reading `.length` unconditionally. Legacy system-keychain credentials display `System Keychain`; this display compatibility does not enable an Antigravity runtime.
 
-Chat model presets: the model `<select>` lists the served subset of `QgridSupportedModel` grouped by provider (`MODEL_PRESET_GROUPS` in `ChatWidget.tsx`), newest first, and defaults to `anthropic/claude-fable-5-1`. Ids that the ChatGPT-subscription Codex route no longer serves (`gpt-5.4` and `gpt-5.4-mini` retired 2026-08-31 with `gpt-5.6-terra` / `gpt-5.6-luna` as replacements; `gpt-5.2` and `gpt-5.3-codex` removed earlier), `gpt-5.3-codex-spark` (rejected on the ChatGPT-account Codex route and absent from its catalog as of 2026-09-29), `gpt-5.5` (retires from the Codex/ChatGPT route on 2026-10-14; replacements `gpt-6.1-sol` / `gpt-6-luna`), the retired `claude-opus-4` / `claude-sonnet-4` (2026-06-15) and `claude-opus-4-1` (2026-08-05; Claude Code silently serves `claude-opus-5-5` instead), and the catalog-less `claude-sonnet-4-7` stay in the SDK type and price table (legacy request-log cost) but are excluded here. web does not depend on ai-sdk, so a new model must be added in both places. `openai/gpt-6-astra`, `openai/gpt-6.1-sol`, `openai/gpt-6-sol`, and `openai/gpt-6-luna` are included in the OpenAI group and `anthropic/claude-opus-5-5` and `anthropic/claude-sonnet-5-5` in the Anthropic group; adding them does not change the default model.
+Chat model presets: the model `<select>` lists the served subset of `QgridSupportedModel` grouped by provider (`MODEL_PRESET_GROUPS` in `ChatWidget.tsx`), newest first, and defaults to `anthropic/claude-fable-5-1`. Ids that the ChatGPT-subscription Codex route no longer serves (`gpt-5.4` and `gpt-5.4-mini` retired 2026-08-31 with `gpt-5.6-terra` / `gpt-5.6-luna` as replacements; `gpt-5.2` and `gpt-5.3-codex` removed earlier), `gpt-5.3-codex-spark` (rejected on the ChatGPT-account Codex route and absent from its catalog as of 2026-09-29), `gpt-5.5` (retires from the Codex/ChatGPT route on 2026-10-14; replacements `gpt-6.1-sol` / `gpt-6-luna`), the retired `claude-opus-4` / `claude-sonnet-4` (2026-06-15) and `claude-opus-4-1` (2026-08-05; Claude Code silently serves `claude-opus-5-5` instead), `claude-sonnet-4-5` (deprecated 2026-09-30, retirement 2026-11-30, replacement `claude-sonnet-5-5`), and the catalog-less `claude-sonnet-4-7` stay in the SDK type and price table (legacy request-log cost) but are excluded here. web does not depend on ai-sdk, so a new model must be added in both places. `openai/gpt-6-astra`, `openai/gpt-6.1-sol`, `openai/gpt-6-sol`, and `openai/gpt-6-luna` are included in the OpenAI group and `anthropic/claude-opus-5-5`, `anthropic/claude-sonnet-5-5`, and `anthropic/claude-haiku-5-5` in the Anthropic group; adding them does not change the default model.
 
 The monit tab's vitals strip shows per-provider in-flight counts, an in-flight sparkline, token chips with quota gauges (fed by the dispatcher's 60-second rate-limits cache; `resetsAt` arrives as ms epoch), and a 1-hour per-provider request/error/cache-hit stats line polled separately every 30 seconds.
 

package/skills/qgrid/references/token-auth-quota-lifecycle.md

@@ -42,6 +42,8 @@ Use this reference before changing token storage, OAuth flows, token activation,
 On creation, the database defaults `reauth_required = false`; `TokenModel.save` also applies `quota_threshold = 80` and `weight = 1` independently for any field not provided (skipped entirely when `id` is present).
 `TokenModel.save` is internal only; callers use the curated qgrid token APIs so they cannot write lifecycle fields directly.
 
+Account re-login reads, deletes, and replaces matching token rows in one explicit Puri transaction. A failed insert rolls back deletion. The oldest matching row supplies quota threshold (including null), weight, and dashboard order unless the caller explicitly supplies them; keepalive stays enabled if any replaced row enabled it. The new login still receives a new ID and resets active/reauth state. Internal full-pool reads use unlimited list mode rather than a 100-row page.
+
 Dashboard credential previews must tolerate legacy credentials without `accessToken`. System-keychain credentials, including a remaining Antigravity row, display `System Keychain` rather than crashing on `.length`. This 2.9.8 display fix neither deletes legacy rows nor enables Antigravity provider execution.
 
 OpenAI credentials: