ima2-gen 1.1.21 → 1.1.22

package/lib/agentRuntime.ts

@@ -1,604 +0,0 @@
-import { randomBytes } from "node:crypto";
-import { mkdir, readFile, unlink, writeFile } from "node:fs/promises";
-import { join } from "node:path";
-import { ulid } from "ulid";
-import { embedImageMetadataBestEffort } from "./imageMetadataStore.js";
-import { invalidateHistoryIndex } from "./historyIndex.js";
-import { logEvent } from "./logger.js";
-import { detectImageMimeFromB64 } from "./refs.js";
-import { resolveProviderOptions } from "./providerOptions.js";
-import { generateViaResponses } from "./responsesImageAdapter.js";
-import { generateViaGrok, type GrokReferenceImage } from "./grokImageAdapter.js";
-import { generateVideoViaGrok } from "./grokVideoAdapter.js";
-import { parseVideoParams } from "./agentGenerationPlanner.js";
-import {
-  appendAgentTurn,
-  buildImageContextManifest,
-  getAgentImages,
-  getAgentSession,
-  importAgentImage,
-  recordAgentWebFinding,
-  restartAgentRuntimeSession,
-} from "./agentStore.js";
-import {
-  AGENT_ALLOWED_TOOLS,
-  type AgentGenerationPlan,
-  type AgentToolCallSummary,
-  type AgentToolName,
-} from "./agentTypes.js";
-import { errInfo } from "./errInfo.js";
-import { type RuntimeContext } from "./runtimeContext.js";
-
-type AgentRunOptions = {
-  provider?: string;
-  quality?: string;
-  size?: string;
-  format?: string;
-  moderation?: string;
-  model?: string;
-  reasoningEffort?: string;
-  requestId?: string;
-  webSearchEnabled?: boolean;
-  parallelism?: number;
-  signal?: AbortSignal | null;
-};
-
-export function assertAgentAllowedTools(tools: readonly string[]) {
-  const allowed = new Set<string>(AGENT_ALLOWED_TOOLS);
-  const denied = tools.filter((tool) => !allowed.has(tool));
-  if (denied.length > 0) {
-    const err = new Error(`Agent tool is not allowed: ${denied.join(", ")}`) as Error & {
-      code?: string;
-      status?: number;
-      deniedTools?: string[];
-    };
-    err.code = "AGENT_TOOL_NOT_ALLOWED";
-    err.status = 403;
-    err.deniedTools = denied;
-    throw err;
-  }
-}
-
-export function agentAllowedToolPayload() {
-  return { tools: [...AGENT_ALLOWED_TOOLS] };
-}
-
-export async function runAgentTurn(ctx: RuntimeContext, sessionId: string, prompt: string, options: AgentRunOptions = {}) {
-  return runAgentGenerationPlan(
-    ctx,
-    sessionId,
-    prompt,
-    {
-      mode: "single",
-      prompts: [prompt],
-      requestedVariants: 1,
-      plannedVariants: 1,
-      plannedParallelism: cleanParallelism(options.parallelism),
-      source: "auto-default",
-      reason: "Direct turn endpoint defaults to one image.",
-      command: null,
-      assistantText: null,
-    },
-    options,
-    { appendUserTurn: true },
-  );
-}
-
-export async function runAgentGenerationPlan(
-  ctx: RuntimeContext,
-  sessionId: string,
-  prompt: string,
-  plan: AgentGenerationPlan,
-  options: AgentRunOptions = {},
-  behavior: { appendUserTurn?: boolean } = {},
-) {
-  const session = getAgentSession(sessionId);
-  if (!session) throw notFound(sessionId);
-  const webSearchEnabled = options.provider === "grok" ? true : options.webSearchEnabled ?? session.webSearchEnabled;
-  const enabledTools: AgentToolName[] = webSearchEnabled
-    ? [...AGENT_ALLOWED_TOOLS]
-    : ["ima2.get_image_context", "ima2.generate_image", "ima2.generate_video"];
-  assertAgentAllowedTools(enabledTools);
-  if (behavior.appendUserTurn !== false) {
-    appendAgentTurn({ sessionId, role: "user", text: prompt, status: "complete" });
-  }
-  if (plan.mode === "question") {
-    const assistantTurn = appendAgentTurn({
-      sessionId,
-      role: "assistant",
-      text: plan.assistantText || plan.reason || "What would you like to clarify before generating images?",
-      imageIds: [],
-      webFindingIds: [],
-      status: "complete",
-    });
-    return { assistantTurn, imageIds: [], webFindingIds: [] };
-  }
-  if (plan.mode === "video") {
-    return runAgentVideoGeneration(ctx, sessionId, prompt, {
-      ...options,
-      requestId: options.requestId ?? `agent_video_${ulid()}`,
-      skipUserTurn: true,
-    });
-  }
-  const manifest = buildImageContextManifest(sessionId);
-  const contextStartedAt = Date.now();
-  appendAgentTurn({
-    sessionId,
-    role: "tool",
-    text: "ima2.get_image_context",
-    status: "complete",
-    raw: {
-      toolCalls: [{
-        id: `tc_context_${ulid()}`,
-        name: "ima2.get_image_context",
-        status: "complete",
-        startedAt: contextStartedAt,
-        finishedAt: Date.now(),
-        durationMs: Date.now() - contextStartedAt,
-        outputSummary: "Loaded current image context manifest.",
-      } satisfies AgentToolCallSummary],
-    },
-  });
-  const generationPrompts = plan.prompts.length > 0 ? plan.prompts : [prompt];
-  const baseRequestId = options.requestId ?? `agent_${ulid()}`;
-  const generationResults = await mapWithLimit(generationPrompts, cleanParallelism(plan.plannedParallelism ?? options.parallelism), async (generationPrompt, index) => {
-    const requestId = generationPrompts.length > 1 ? `${baseRequestId}_${index + 1}` : baseRequestId;
-    const startedAt = Date.now();
-    const result = await runGeneratorWithRuntimeRecovery(ctx, sessionId, generationPrompt, manifest, webSearchEnabled, {
-      ...options,
-      requestId,
-    });
-    const findingIds = recordSearchFindings(sessionId, generationPrompt, result.webSearchCalls, result.provider ?? "oauth");
-    const finishedAt = Date.now();
-    return {
-      prompt: generationPrompt,
-      imageId: result.image.id,
-      text: result.text,
-      findingIds,
-      toolCall: {
-        id: `tc_generate_${ulid()}`,
-        name: "ima2.generate_image",
-        status: "complete",
-        startedAt,
-        finishedAt,
-        durationMs: finishedAt - startedAt,
-        requestId,
-        inputSummary: generationPrompt,
-        outputSummary: `Generated ${result.image.filename}. ${plan.reason}`,
-        imageIds: [result.image.id],
-        webFindingIds: findingIds,
-      } satisfies AgentToolCallSummary,
-    };
-  });
-  const imageIds = generationResults.map((result) => result.imageId);
-  const responseTexts = generationResults
-    .map((result) => result.text)
-    .filter((text): text is string => typeof text === "string" && text.trim().length > 0);
-  const findingIds = generationResults.flatMap((result) => result.findingIds);
-  const webToolCall: AgentToolCallSummary | null = webSearchEnabled ? {
-    id: `tc_web_${ulid()}`,
-    name: "ima2.web_search",
-    status: "complete",
-    outputSummary: findingIds.length > 0
-      ? `Recorded ${findingIds.length} web finding${findingIds.length === 1 ? "" : "s"}.`
-      : "Web search enabled; no findings were reported.",
-    webFindingIds: findingIds,
-  } : null;
-  appendAgentTurn({
-    sessionId,
-    role: "tool",
-    text: webSearchEnabled ? "ima2.web_search + ima2.generate_image" : "ima2.generate_image",
-    imageIds,
-    webFindingIds: findingIds,
-    status: "complete",
-    raw: {
-      toolCalls: [
-        ...(webToolCall ? [webToolCall] : []),
-        ...generationResults.map((result) => result.toolCall),
-      ],
-    },
-  });
-  const assistantTurn = appendAgentTurn({
-    sessionId,
-    role: "assistant",
-    text: formatAgentAssistantText(plan, imageIds.length, responseTexts),
-    imageIds,
-    webFindingIds: findingIds,
-    status: "complete",
-  });
-  return { assistantTurn, imageIds, webFindingIds: findingIds };
-}
-
-function formatAgentAssistantText(plan: AgentGenerationPlan, imageCount: number, responseTexts: readonly string[]): string {
-  const countText = imageCount === 1 ? "Generated 1 image artifact." : `Generated ${imageCount} image artifacts.`;
-  const modeText = plan.mode === "fanout"
-    ? `Fanout used ${plan.plannedParallelism} concurrent tool call${plan.plannedParallelism === 1 ? "" : "s"}.`
-    : "Single-image plan completed.";
-  const modelText = responseTexts.length > 0 ? `${responseTexts.join("\n\n")}\n\n` : "";
-  return `${modelText}${countText} ${modeText} ${plan.reason}`.trim();
-}
-
-async function runGeneratorWithRuntimeRecovery(
-  ctx: RuntimeContext,
-  sessionId: string,
-  prompt: string,
-  manifest: string,
-  webSearchEnabled: boolean,
-  options: AgentRunOptions,
-) {
-  try {
-    return await generateAgentImageWithRetry(ctx, sessionId, prompt, manifest, webSearchEnabled, options);
-  } catch (error) {
-    const err = errInfo(error);
-    if (isRuntimeRestartableError(error)) {
-      restartAgentRuntimeSession(sessionId, err.code || err.message);
-    }
-    appendAgentTurn({ sessionId, role: "assistant", text: err.message, status: "error" });
-    throw error;
-  }
-}
-
-export function isRuntimeRestartableError(error: unknown) {
-  const err = errInfo(error);
-  const code = err.code || "";
-  return (
-    code.includes("AUTH") ||
-    code.includes("TIMEOUT") ||
-    code.includes("PROTOCOL") ||
-    err.message.toLowerCase().includes("protocol wedge")
-  );
-}
-
-async function generateAgentImageWithRetry(
-  ctx: RuntimeContext,
-  sessionId: string,
-  prompt: string,
-  manifest: string,
-  webSearchEnabled: boolean,
-  options: AgentRunOptions,
-) {
-  let lastError: unknown = null;
-  for (let attempt = 0; attempt < 2; attempt++) {
-    try {
-      const forcedPrompt = attempt === 0 ? prompt : forceImagePrompt(prompt);
-      const result = await generateAgentImage(ctx, sessionId, forcedPrompt, manifest, webSearchEnabled, options);
-      if (result.image) return result;
-    } catch (error) {
-      lastError = error;
-      if (!isTextOnlyResult(error)) throw error;
-      if (attempt === 1) break;
-      appendAgentTurn({
-        sessionId,
-        role: "tool",
-        text: "ima2.generate_image retry: text-only result rejected",
-        status: "error",
-      });
-    }
-  }
-  throw textOnlyError(lastError);
-}
-
-async function generateAgentImage(
-  ctx: RuntimeContext,
-  sessionId: string,
-  prompt: string,
-  manifest: string,
-  webSearchEnabled: boolean,
-  options: AgentRunOptions,
-) {
-  const requestId = options.requestId ?? `agent_${ulid()}`;
-  const providerOptions = resolveProviderOptions(ctx, {
-    provider: options.provider ?? "oauth",
-    rawModel: options.model,
-    rawReasoningEffort: options.reasoningEffort,
-    rawSize: options.size ?? "1024x1024",
-    rawWebSearchEnabled: webSearchEnabled,
-    searchMode: webSearchEnabled ? "on" : "off",
-  });
-  if (providerOptions.error) {
-    const err = new Error(providerOptions.error) as Error & { code?: string; status?: number };
-    err.code = providerOptions.code;
-    err.status = providerOptions.status;
-    throw err;
-  }
-  const activeProvider = providerOptions.provider;
-  const effectiveModel = activeProvider === "grok" && options.quality === "high"
-    ? "grok-imagine-image-quality"
-    : providerOptions.model;
-  const response = activeProvider === "grok"
-    ? await generateViaGrok(`${manifest}\n\nUser request:\n${prompt}`, ctx, {
-        model: effectiveModel,
-        size: providerOptions.size,
-        requestId,
-        signal: options.signal ?? undefined,
-        references: await loadAgentCurrentImageReferences(ctx, sessionId),
-      })
-    : await generateViaResponses(
-        activeProvider,
-        `${manifest}\n\nUser request:\n${prompt}`,
-        options.quality ?? "medium",
-        providerOptions.size,
-        options.moderation ?? "low",
-        [],
-        requestId,
-        "auto",
-        ctx,
-        {
-          model: providerOptions.model,
-          reasoningEffort: providerOptions.reasoningEffort,
-          webSearchEnabled,
-          signal: options.signal,
-        },
-      );
-  const format = activeProvider === "grok"
-    ? imageFormatFromMime(("mime" in response ? response.mime : undefined) || detectImageMimeFromB64(response.b64) || "image/jpeg")
-    : options.format ?? "png";
-  const image = await persistAgentImage(ctx, sessionId, prompt, format, requestId, response, {
-    provider: String(activeProvider),
-    model: String(effectiveModel),
-  });
-  const responseText = "text" in response && typeof response.text === "string" ? response.text : null;
-  return { image, webSearchCalls: response.webSearchCalls || 0, text: responseText, provider: activeProvider };
-}
-
-async function loadAgentCurrentImageReferences(ctx: RuntimeContext, sessionId: string): Promise<GrokReferenceImage[]> {
-  const session = getAgentSession(sessionId);
-  const currentImage = session?.lastImageId
-    ? getAgentImages(sessionId).find((image) => image.id === session.lastImageId)
-    : null;
-  if (!currentImage?.filename) return [];
-  try {
-    const b64 = (await readFile(join(ctx.config.storage.generatedDir, currentImage.filename))).toString("base64");
-    const mime = detectImageMimeFromB64(b64);
-    return [{ b64, declaredMime: mime, detectedMime: mime }];
-  } catch (error) {
-    const err = errInfo(error);
-    logEvent("agent", "grok_ref_missing", { sessionId, filename: currentImage.filename, code: err.code, message: err.message });
-    return [];
-  }
-}
-
-function imageFormatFromMime(mime: string | null | undefined): "png" | "jpeg" | "webp" {
-  if (mime === "image/jpeg") return "jpeg";
-  if (mime === "image/webp") return "webp";
-  return "png";
-}
-
-async function persistAgentImage(
-  ctx: RuntimeContext,
-  sessionId: string,
-  prompt: string,
-  format: string,
-  requestId: string,
-  response: { b64: string; revisedPrompt?: string | null; usage?: unknown; webSearchCalls?: number },
-  generation: { provider: string; model: string },
-) {
-  await mkdir(ctx.config.storage.generatedDir, { recursive: true });
-  const rand = randomBytes(ctx.config.ids.generatedHexBytes).toString("hex");
-  const filename = `${Date.now()}_${rand}_agent.${format}`;
-  const meta = {
-    kind: "agent",
-    requestId,
-    sessionId,
-    prompt,
-    userPrompt: prompt,
-    revisedPrompt: response.revisedPrompt ?? null,
-    provider: generation.provider,
-    model: generation.model,
-    createdAt: Date.now(),
-    usage: response.usage ?? null,
-    webSearchCalls: response.webSearchCalls ?? 0,
-  };
-  const embedded = await embedImageMetadataBestEffort(Buffer.from(response.b64, "base64"), format, meta, {
-    version: ctx.packageVersion,
-  });
-  const filePath = join(ctx.config.storage.generatedDir, filename);
-  await writeFile(filePath, embedded.buffer);
-  try {
-    await writeFile(`${filePath}.json`, JSON.stringify(meta));
-  } catch (err) {
-    await unlink(filePath).catch(() => {});
-    throw err;
-  }
-  invalidateHistoryIndex();
-  logEvent("agent", "saved", { requestId, sessionId, filename });
-  return importAgentImage(sessionId, {
-    id: `ai_${ulid()}`,
-    filename,
-    url: `/generated/${filename}`,
-    prompt,
-    revisedPrompt: response.revisedPrompt ?? null,
-    createdAt: Date.now(),
-  });
-}
-
-export async function runAgentVideoGeneration(
-  ctx: RuntimeContext,
-  sessionId: string,
-  prompt: string,
-  options: AgentRunOptions & { skipUserTurn?: boolean } = {},
-) {
-  const session = getAgentSession(sessionId);
-  if (!session) throw notFound(sessionId);
-  if (!options.skipUserTurn) {
-    appendAgentTurn({ sessionId, role: "user", text: prompt, status: "complete" });
-  }
-  const requestId = options.requestId ?? `agent_video_${ulid()}`;
-  const startedAt = Date.now();
-
-  // Auto I2V: if session has a last image, use it as source
-  let sourceImage: string | undefined;
-  let mode: "text-to-video" | "image-to-video" = "text-to-video";
-  if (session.lastImageId) {
-    const images = getAgentImages(sessionId);
-    const lastImage = images.find((img) => img.id === session.lastImageId);
-    if (lastImage?.filename && !lastImage.filename.endsWith(".mp4")) {
-      try {
-        const { loadAssetB64 } = await import("./nodeStore.js");
-        sourceImage = await loadAssetB64(ctx.rootDir, lastImage.filename, ctx.config.storage.generatedDir);
-        mode = "image-to-video";
-      } catch { /* fallback to T2V */ }
-    }
-  }
-
-  const videoParams = parseVideoParams(prompt);
-
-  const result = await generateVideoViaGrok(prompt, ctx, {
-    model: "grok-imagine-video",
-    mode,
-    sourceImage,
-    duration: videoParams.duration ?? 5,
-    resolution: videoParams.resolution ?? "480p",
-    aspectRatio: (videoParams.aspectRatio ?? "auto") as "auto" | "1:1" | "16:9" | "9:16" | "4:3" | "3:4" | "3:2" | "2:3",
-    requestId,
-    signal: options.signal ?? undefined,
-  });
-  const video = await persistAgentVideo(ctx, sessionId, prompt, requestId, result);
-  const finishedAt = Date.now();
-  const toolCall: AgentToolCallSummary = {
-    id: `tc_video_${ulid()}`,
-    name: "ima2.generate_video",
-    status: "complete",
-    startedAt,
-    finishedAt,
-    durationMs: finishedAt - startedAt,
-    requestId,
-    inputSummary: prompt,
-    outputSummary: `Generated video ${video.filename}.`,
-    imageIds: [video.id],
-  };
-  appendAgentTurn({
-    sessionId,
-    role: "tool",
-    text: "ima2.generate_video",
-    imageIds: [video.id],
-    status: "complete",
-    raw: { toolCalls: [toolCall] },
-  });
-  const assistantTurn = appendAgentTurn({
-    sessionId,
-    role: "assistant",
-    text: `Generated 1 video artifact. ${result.revisedPrompt}`,
-    imageIds: [video.id],
-    status: "complete",
-  });
-  return { assistantTurn, imageIds: [video.id], webFindingIds: [] };
-}
-
-async function persistAgentVideo(
-  ctx: RuntimeContext,
-  sessionId: string,
-  prompt: string,
-  requestId: string,
-  result: { videoBuffer: Buffer; revisedPrompt: string; usage: Record<string, number> | null; webSearchCalls: number },
-) {
-  await mkdir(ctx.config.storage.generatedDir, { recursive: true });
-  const rand = randomBytes(ctx.config.ids.generatedHexBytes).toString("hex");
-  const filename = `${Date.now()}_${rand}_agent.mp4`;
-  const meta = {
-    kind: "agent",
-    mediaType: "video",
-    requestId,
-    sessionId,
-    prompt,
-    userPrompt: prompt,
-    revisedPrompt: result.revisedPrompt,
-    provider: "grok",
-    model: "grok-imagine-video",
-    createdAt: Date.now(),
-    usage: result.usage,
-    webSearchCalls: result.webSearchCalls,
-  };
-  const filePath = join(ctx.config.storage.generatedDir, filename);
-  await writeFile(filePath, result.videoBuffer);
-  try {
-    await writeFile(`${filePath}.json`, JSON.stringify(meta));
-  } catch (err) {
-    await unlink(filePath).catch(() => {});
-    throw err;
-  }
-  invalidateHistoryIndex();
-  logEvent("agent", "video_saved", { requestId, sessionId, filename });
-  return importAgentImage(sessionId, {
-    id: `ai_${ulid()}`,
-    filename,
-    url: `/generated/${filename}`,
-    prompt,
-    revisedPrompt: result.revisedPrompt,
-    createdAt: Date.now(),
-  });
-}
-
-function recordSearchFindings(sessionId: string, prompt: string, count: number, provider: string) {
-  if (!count) return [];
-  const isGrok = provider === "grok";
-  return [
-    recordAgentWebFinding({
-      sessionId,
-      query: prompt,
-      title: isGrok ? "Grok visual research" : "Responses web_search",
-      snippet: `${isGrok ? "Grok" : "Responses"} reported ${count} web search call${count === 1 ? "" : "s"}.`,
-    }),
-  ];
-}
-
-function forceImagePrompt(prompt: string) {
-  return [
-    "The previous turn did not return an image artifact.",
-    "Return a final image using ima2.generate_image/image_generation now.",
-    `User request: ${prompt}`,
-  ].join("\n");
-}
-
-function isTextOnlyResult(error: unknown) {
-  const err = errInfo(error);
-  return [
-    "EMPTY_RESPONSE",
-    "IMAGE_TOOL_NOT_CALLED",
-    "WEB_SEARCH_ONLY_RESPONSE",
-    "IMAGE_TOOL_COMPLETED_WITHOUT_RESULT",
-  ].includes(err.code || "") || err.message.includes("No image data");
-}
-
-function textOnlyError(cause: unknown) {
-  const err = new Error("Agent result did not include an image artifact.") as Error & {
-    code?: string;
-    status?: number;
-    cause?: unknown;
-  };
-  err.code = "AGENT_TEXT_ONLY_RESULT";
-  err.status = 422;
-  err.cause = cause;
-  return err;
-}
-
-async function mapWithLimit<T, R>(
-  items: readonly T[],
-  limit: number,
-  mapper: (item: T, index: number) => Promise<R>,
-): Promise<R[]> {
-  const results: R[] = [];
-  let nextIndex = 0;
-  const workerCount = Math.min(limit, items.length);
-  await Promise.all(Array.from({ length: workerCount }, async () => {
-    while (nextIndex < items.length) {
-      const currentIndex = nextIndex;
-      nextIndex += 1;
-      results[currentIndex] = await mapper(items[currentIndex], currentIndex);
-    }
-  }));
-  return results;
-}
-
-function cleanParallelism(value: unknown) {
-  const numeric = typeof value === "number" ? value : Number(value);
-  if (!Number.isFinite(numeric)) return 2;
-  return Math.max(1, Math.min(8, Math.round(numeric)));
-}
-
-function notFound(sessionId: string) {
-  const err = new Error(`Agent session not found: ${sessionId}`) as Error & { code?: string; status?: number };
-  err.code = "AGENT_SESSION_NOT_FOUND";
-  err.status = 404;
-  return err;
-}

package/lib/agentSettings.ts

@@ -1,72 +0,0 @@
-import type { AgentGenerationSettings } from "./agentTypes.js";
-
-const PROVIDERS = new Set(["oauth", "api", "grok"]);
-const QUALITIES = new Set(["low", "medium", "high"]);
-const FORMATS = new Set(["png", "jpeg", "webp"]);
-const MODERATIONS = new Set(["auto", "low"]);
-const REASONING_EFFORTS = new Set(["low", "medium", "high", "xhigh"]);
-const GENERATION_STRATEGIES = new Set(["auto", "manual"]);
-
-export const DEFAULT_AGENT_GENERATION_SETTINGS: AgentGenerationSettings = {
-  provider: "oauth",
-  model: "gpt-5.4-mini",
-  quality: "medium",
-  size: "1024x1024",
-  format: "png",
-  moderation: "low",
-  reasoningEffort: "medium",
-  webSearchEnabled: true,
-  generationStrategy: "auto",
-  variants: 1,
-  maxAutoVariants: 8,
-  parallelism: 2,
-};
-
-export function normalizeAgentGenerationSettings(
-  value: unknown,
-  fallback: AgentGenerationSettings = DEFAULT_AGENT_GENERATION_SETTINGS,
-): AgentGenerationSettings {
-  const input = value && typeof value === "object" ? value as Record<string, unknown> : {};
-  return {
-    provider: cleanEnum(input.provider, PROVIDERS, fallback.provider),
-    model: cleanString(input.model, fallback.model),
-    quality: cleanEnum(input.quality, QUALITIES, fallback.quality),
-    size: cleanSize(input.size, fallback.size),
-    format: cleanEnum(input.format, FORMATS, fallback.format),
-    moderation: cleanEnum(input.moderation, MODERATIONS, fallback.moderation),
-    reasoningEffort: cleanEnum(input.reasoningEffort, REASONING_EFFORTS, fallback.reasoningEffort),
-    webSearchEnabled: typeof input.webSearchEnabled === "boolean" ? input.webSearchEnabled : fallback.webSearchEnabled,
-    generationStrategy: cleanEnum(input.generationStrategy, GENERATION_STRATEGIES, fallback.generationStrategy),
-    variants: cleanPositiveInt(input.variants, fallback.variants, 1, 8),
-    maxAutoVariants: cleanPositiveInt(input.maxAutoVariants, fallback.maxAutoVariants, 1, 8),
-    parallelism: cleanPositiveInt(input.parallelism, fallback.parallelism, 1, 8),
-  };
-}
-
-export function mergeAgentGenerationSettings(
-  current: AgentGenerationSettings,
-  patch: unknown,
-): AgentGenerationSettings {
-  return normalizeAgentGenerationSettings({ ...current, ...(patch && typeof patch === "object" ? patch : {}) }, current);
-}
-
-function cleanEnum<T extends string>(value: unknown, allowed: Set<string>, fallback: T): T {
-  return typeof value === "string" && allowed.has(value) ? value as T : fallback;
-}
-
-function cleanString(value: unknown, fallback: string): string {
-  return typeof value === "string" && value.trim() ? value.trim().slice(0, 120) : fallback;
-}
-
-function cleanSize(value: unknown, fallback: string): string {
-  if (typeof value !== "string") return fallback;
-  const trimmed = value.trim();
-  if (trimmed === "auto") return trimmed;
-  return /^\d{3,4}x\d{3,4}$/.test(trimmed) ? trimmed : fallback;
-}
-
-function cleanPositiveInt(value: unknown, fallback: number, min: number, max: number): number {
-  const numeric = typeof value === "number" ? value : Number(value);
-  if (!Number.isFinite(numeric)) return fallback;
-  return Math.max(min, Math.min(max, Math.round(numeric)));
-}