mulmocast 1.2.3 → 1.2.5

package/lib/actions/audio.d.ts

@@ -1,6 +1,5 @@
 import "dotenv/config";
-import type { CallbackFunction } from "graphai";
-import { MulmoStudioContext, MulmoBeat } from "../types/index.js";
+import { MulmoStudioContext, MulmoBeat, PublicAPIArgs } from "../types/index.js";
 export declare const getBeatAudioPath: (text: string, context: MulmoStudioContext, beat: MulmoBeat, lang?: string) => string | undefined;
-export declare const generateBeatAudio: (index: number, context: MulmoStudioContext, settings?: Record<string, string>, callbacks?: CallbackFunction[]) => Promise<void>;
-export declare const audio: (context: MulmoStudioContext, settings?: Record<string, string>, callbacks?: CallbackFunction[]) => Promise<MulmoStudioContext>;
+export declare const generateBeatAudio: (index: number, context: MulmoStudioContext, args?: PublicAPIArgs) => Promise<void>;
+export declare const audio: (context: MulmoStudioContext, args?: PublicAPIArgs) => Promise<MulmoStudioContext>;

package/lib/actions/audio.js

@@ -186,7 +186,8 @@ const audioAgents = {
     addBGMAgent,
     combineAudioFilesAgent,
 };
-export const generateBeatAudio = async (index, context, settings, callbacks) => {
+export const generateBeatAudio = async (index, context, args) => {
+    const { settings, callbacks } = args ?? {};
     try {
         MulmoStudioContextMethods.setSessionState(context, "audio", true);
         const fileName = MulmoStudioContextMethods.getFileName(context);
@@ -214,7 +215,8 @@ export const generateBeatAudio = async (index, context, settings, callbacks) =>
         MulmoStudioContextMethods.setSessionState(context, "audio", false);
     }
 };
-export const audio = async (context, settings, callbacks) => {
+export const audio = async (context, args) => {
+    const { settings, callbacks } = args ?? {};
     try {
         MulmoStudioContextMethods.setSessionState(context, "audio", true);
         const fileName = MulmoStudioContextMethods.getFileName(context);

package/lib/actions/captions.d.ts

@@ -1,3 +1,2 @@
-import { MulmoStudioContext } from "../types/index.js";
-import type { CallbackFunction } from "graphai";
-export declare const captions: (context: MulmoStudioContext, callbacks?: CallbackFunction[]) => Promise<MulmoStudioContext>;
+import { MulmoStudioContext, PublicAPIArgs } from "../types/index.js";
+export declare const captions: (context: MulmoStudioContext, args?: PublicAPIArgs) => Promise<MulmoStudioContext>;

package/lib/actions/captions.js

@@ -69,7 +69,8 @@ const graph_data = {
         },
     },
 };
-export const captions = async (context, callbacks) => {
+export const captions = async (context, args) => {
+    const { callbacks } = args ?? {};
     if (MulmoStudioContextMethods.getCaption(context)) {
         try {
             MulmoStudioContextMethods.setSessionState(context, "caption", true);

package/lib/actions/images.d.ts

@@ -1,20 +1,18 @@
-import type { GraphOptions, CallbackFunction } from "graphai";
-import { MulmoStudioContext } from "../types/index.js";
+import type { GraphOptions } from "graphai";
+import { MulmoStudioContext, PublicAPIArgs } from "../types/index.js";
 export declare const graphOption: (context: MulmoStudioContext, settings?: Record<string, string>) => Promise<GraphOptions>;
 type ImageOptions = {
     imageAgents: Record<string, unknown>;
 };
-export declare const images: (context: MulmoStudioContext, args?: {
-    settings?: Record<string, string>;
-    callbacks?: CallbackFunction[];
+export declare const images: (context: MulmoStudioContext, args?: PublicAPIArgs & {
     options?: ImageOptions;
 }) => Promise<MulmoStudioContext>;
 export declare const generateBeatImage: (inputs: {
     index: number;
     context: MulmoStudioContext;
-    settings?: Record<string, string>;
-    callbacks?: CallbackFunction[];
-    forceMovie?: boolean;
-    forceImage?: boolean;
+    args?: PublicAPIArgs & {
+        forceMovie?: boolean;
+        forceImage?: boolean;
+    };
 }) => Promise<void>;
 export {};

package/lib/actions/images.js

@@ -346,8 +346,8 @@ export const graphOption = async (context, settings) => {
             },
         ],
         taskManager: new TaskManager(MulmoPresentationStyleMethods.getConcurrency(context.presentationStyle)),
+        config: settings2GraphAIConfig(settings, process.env),
     };
-    options.config = settings2GraphAIConfig(settings, process.env);
     return options;
 };
 const prepareGenerateImages = async (context) => {
@@ -367,7 +367,8 @@ const prepareGenerateImages = async (context) => {
     };
     return injections;
 };
-const generateImages = async (context, settings, callbacks, options) => {
+const generateImages = async (context, args) => {
+    const { settings, callbacks, options } = args ?? {};
     const optionImageAgents = options?.imageAgents ?? {};
     const injections = await prepareGenerateImages(context);
     const graphaiAgent = {
@@ -388,10 +389,9 @@ const generateImages = async (context, settings, callbacks, options) => {
 };
 // public api
 export const images = async (context, args) => {
-    const { settings, callbacks, options } = args ?? {};
     try {
         MulmoStudioContextMethods.setSessionState(context, "image", true);
-        const newContext = await generateImages(context, settings, callbacks, options);
+        const newContext = await generateImages(context, args);
         MulmoStudioContextMethods.setSessionState(context, "image", false);
         return newContext;
     }
@@ -402,7 +402,8 @@ export const images = async (context, args) => {
 };
 // public api
 export const generateBeatImage = async (inputs) => {
-    const { index, context, settings, callbacks, forceMovie, forceImage } = inputs;
+    const { index, context, args } = inputs;
+    const { settings, callbacks, forceMovie, forceImage } = args ?? {};
     const options = await graphOption(context, settings);
     const injections = await prepareGenerateImages(context);
     const graph = new GraphAI(beat_graph_data, defaultAgents, options);

package/lib/actions/translate.d.ts

@@ -1,6 +1,5 @@
 import "dotenv/config";
-import type { CallbackFunction } from "graphai";
-import type { LANG, MulmoStudioContext } from "../types/index.js";
+import type { LANG, MulmoStudioContext, PublicAPIArgs } from "../types/index.js";
 export declare const translateTextGraph: {
     version: number;
     nodes: {
@@ -50,11 +49,5 @@ export declare const getOutputMultilingualFilePathAndMkdir: (context: MulmoStudi
     outputMultilingualFilePath: string;
     outDirPath: string;
 };
-export declare const translateBeat: (index: number, context: MulmoStudioContext, targetLangs: string[], args?: {
-    settings?: Record<string, string>;
-    callbacks?: CallbackFunction[];
-}) => Promise<void>;
-export declare const translate: (context: MulmoStudioContext, args?: {
-    callbacks?: CallbackFunction[];
-    settings?: Record<string, string>;
-}) => Promise<MulmoStudioContext>;
+export declare const translateBeat: (index: number, context: MulmoStudioContext, targetLangs: string[], args?: PublicAPIArgs) => Promise<void>;
+export declare const translate: (context: MulmoStudioContext, args?: PublicAPIArgs) => Promise<MulmoStudioContext>;

package/lib/agents/image_genai_agent.js

@@ -8,7 +8,7 @@ export const imageGenAIAgent = async ({ namedInputs, params, config, }) => {
     const model = params.model ?? provider2ImageAgent["google"].defaultModel;
     const apiKey = config?.apiKey;
     if (!apiKey) {
-        throw new Error("API key is required for Google GenAI agent");
+        throw new Error("Google GenAI API key is required (GEMINI_API_KEY)");
     }
     try {
         const ai = new GoogleGenAI({ apiKey });

package/lib/agents/image_openai_agent.js

@@ -8,6 +8,9 @@ export const imageOpenaiAgent = async ({ namedInputs, params, config, }) => {
     const { prompt, referenceImages } = namedInputs;
     const { moderation, canvasSize, quality } = params;
     const { apiKey, baseURL } = { ...config };
+    if (!apiKey) {
+        throw new Error("OpenAI API key is required (OPENAI_API_KEY)");
+    }
     const model = params.model ?? provider2ImageAgent["openai"].defaultModel;
     const openai = new OpenAI({ apiKey, baseURL });
     const size = (() => {

package/lib/agents/lipsync_replicate_agent.js

@@ -7,7 +7,7 @@ export const lipSyncReplicateAgent = async ({ namedInputs, params, config, }) =>
     const apiKey = config?.apiKey;
     const model = params.model ?? provider2LipSyncAgent.replicate.defaultModel;
     if (!apiKey) {
-        throw new Error("REPLICATE_API_TOKEN environment variable is required");
+        throw new Error("Replicate API key is required (REPLICATE_API_TOKEN)");
     }
     const replicate = new Replicate({
         auth: apiKey,

package/lib/agents/movie_genai_agent.js

@@ -19,7 +19,7 @@ export const movieGenAIAgent = async ({ namedInputs, params, config, }) => {
     const duration = params.duration ?? 8;
     const apiKey = config?.apiKey;
     if (!apiKey) {
-        throw new Error("API key is required for Google GenAI agent");
+        throw new Error("Google GenAI API key is required (GEMINI_API_KEY)");
     }
     try {
         const ai = new GoogleGenAI({ apiKey });

package/lib/agents/movie_replicate_agent.js

@@ -86,7 +86,7 @@ export const movieReplicateAgent = async ({ namedInputs, params, config, }) => {
     }
     const apiKey = config?.apiKey;
     if (!apiKey) {
-        throw new Error("REPLICATE_API_TOKEN environment variable is required");
+        throw new Error("Replicate API key is required (REPLICATE_API_TOKEN)");
     }
     try {
         const buffer = await generateMovie(model, apiKey, prompt, imagePath, aspectRatio, duration);

package/lib/agents/sound_effect_replicate_agent.js

@@ -7,7 +7,7 @@ export const soundEffectReplicateAgent = async ({ namedInputs, params, config })
     const apiKey = config?.apiKey;
     const model = params.model ?? provider2SoundEffectAgent.replicate.defaultModel;
     if (!apiKey) {
-        throw new Error("REPLICATE_API_TOKEN environment variable is required");
+        throw new Error("Replicate API key is required (REPLICATE_API_TOKEN)");
     }
     const replicate = new Replicate({
         auth: apiKey,

package/lib/agents/tts_elevenlabs_agent.js

@@ -5,7 +5,7 @@ export const ttsElevenlabsAgent = async ({ namedInputs, params, config, }) => {
     const { voice, model, stability, similarityBoost, suppressError } = params;
     const apiKey = config?.apiKey;
     if (!apiKey) {
-        throw new Error("ELEVENLABS_API_KEY environment variable is required");
+        throw new Error("ElevenLabs API key is required (ELEVENLABS_API_KEY)");
     }
     if (!voice) {
         throw new Error("ELEVENLABS Voice ID is required");

package/lib/agents/tts_nijivoice_agent.js

@@ -1,15 +1,19 @@
-import { GraphAILogger, assert } from "graphai";
+import { GraphAILogger } from "graphai";
+/*
 const errorMessage = [
-    "TTS NijiVoice: No API key. ",
-    "You have the following options:",
-    "1. Obtain an API key from Niji Voice (https://platform.nijivoice.com/) and set it as the NIJIVOICE_API_KEY environment variable.",
-    '2. Use OpenAI\'s TTS instead of Niji Voice by changing speechParams.provider from "nijivoice" to "openai".',
+  "TTS NijiVoice: No API key. ",
+  "You have the following options:",
+  "1. Obtain an API key from Niji Voice (https://platform.nijivoice.com/) and set it as the NIJIVOICE_API_KEY environment variable.",
+  '2. Use OpenAI\'s TTS instead of Niji Voice by changing speechParams.provider from "nijivoice" to "openai".',
 ].join("\n");
+*/
 export const ttsNijivoiceAgent = async ({ params, namedInputs, config, }) => {
     const { suppressError, voice, speed, speed_global } = params;
     const { apiKey } = config ?? {};
     const { text } = namedInputs;
-    assert(!!apiKey, errorMessage);
+    if (!apiKey) {
+        throw new Error("NijiVoice API key is required (NIJIVOICE_API_KEY)");
+    }
     const url = `https://api.nijivoice.com/api/platform/v1/voice-actors/${voice}/generate-voice`;
     const options = {
         method: "POST",

package/lib/agents/tts_openai_agent.js

@@ -5,6 +5,9 @@ export const ttsOpenaiAgent = async ({ namedInputs, params, config, }) => {
     const { text } = namedInputs;
     const { model, voice, suppressError, instructions } = params;
     const { apiKey, baseURL } = config ?? {};
+    if (!apiKey) {
+        throw new Error("OpenAI API key is required (OPENAI_API_KEY)");
+    }
     const openai = new OpenAI({ apiKey, baseURL });
     try {
         const tts_options = {

package/lib/types/type.d.ts

@@ -1,3 +1,4 @@
+import { type CallbackFunction } from "graphai";
 import { langSchema, localizedTextSchema, mulmoBeatSchema, mulmoScriptSchema, mulmoStudioSchema, mulmoStudioBeatSchema, mulmoStoryboardSchema, mulmoStoryboardSceneSchema, mulmoStudioMultiLingualSchema, mulmoStudioMultiLingualDataSchema, mulmoStudioMultiLingualFileSchema, speakerDictionarySchema, mulmoImageParamsSchema, mulmoImageParamsImagesSchema, mulmoFillOptionSchema, mulmoMovieParamsSchema, textSlideParamsSchema, speechOptionsSchema, speakerDataSchema, mulmoCanvasDimensionSchema, mulmoPromptTemplateSchema, mulmoPromptTemplateFileSchema, text2ImageProviderSchema, text2HtmlImageProviderSchema, text2MovieProviderSchema, text2SpeechProviderSchema, mulmoPresentationStyleSchema, multiLingualTextsSchema, mulmoImageAssetSchema, mulmoMermaidMediaSchema, mulmoTextSlideMediaSchema, mulmoMarkdownMediaSchema, mulmoImageMediaSchema, mulmoChartMediaSchema, mediaSourceSchema, mulmoSessionStateSchema, mulmoOpenAIImageModelSchema, mulmoGoogleImageModelSchema, mulmoGoogleMovieModelSchema, mulmoReplicateMovieModelSchema, mulmoImagePromptMediaSchema } from "./schema.js";
 import { pdf_modes, pdf_sizes, storyToScriptGenerateMode } from "../utils/const.js";
 import type { LLM } from "../utils/provider2agent.js";
@@ -128,3 +129,7 @@ export type InitOptions = {
     c?: string;
     p?: string;
 };
+export type PublicAPIArgs = {
+    settings?: Record<string, string>;
+    callbacks?: CallbackFunction[];
+};

package/lib/utils/provider2agent.d.ts

@@ -95,26 +95,31 @@ export declare const provider2LLMAgent: {
         readonly agentName: "openAIAgent";
         readonly defaultModel: "gpt-5";
         readonly max_tokens: 8192;
+        readonly models: readonly ["gpt-5", "gpt-5-nano", "gpt-5-mini", "gpt-4.1", "gpt-4.1-mini", "gpt-4.1-nano", "o3", "o3-mini", "o3-pro", "o1", "o1-pro", "gpt-4o", "gpt-4o-mini"];
     };
     readonly anthropic: {
         readonly agentName: "anthropicAgent";
         readonly defaultModel: "claude-3-7-sonnet-20250219";
         readonly max_tokens: 8192;
+        readonly models: readonly ["claude-opus-4-1-20250805", "claude-opus-4-20250514", "claude-sonnet-4-20250514", "claude-3-7-sonnet-20250219", "claude-3-haiku-20240307"];
     };
     readonly gemini: {
         readonly agentName: "geminiAgent";
-        readonly defaultModel: "gemini-1.5-flash";
+        readonly defaultModel: "gemini-2.5-flash";
         readonly max_tokens: 8192;
+        readonly models: readonly ["gemini-2.5-pro", "gemini-2.5-flash", "gemini-2.5-flash-lite", "gemini-2.0-flash"];
     };
     readonly groq: {
         readonly agentName: "groqAgent";
-        readonly defaultModel: "llama3-8b-8192";
+        readonly defaultModel: "llama-3.1-8b-instant";
         readonly max_tokens: 4096;
+        readonly models: readonly ["llama-3.1-8b-instant", "llama-3.3-70b-versatile", "deepseek-r1-distill-llama-70b", "openai/gpt-oss-120b", "openai/gpt-oss-20b"];
     };
     readonly mock: {
         readonly agentName: "mediaMockAgent";
         readonly defaultModel: "mock";
         readonly max_tokens: 4096;
+        readonly models: readonly ["mock"];
     };
 };
 export declare const defaultProviders: {

package/lib/utils/provider2agent.js

@@ -208,26 +208,45 @@ export const provider2LLMAgent = {
         agentName: "openAIAgent",
         defaultModel: "gpt-5",
         max_tokens: 8192,
+        models: [
+            "gpt-5",
+            "gpt-5-nano",
+            "gpt-5-mini",
+            "gpt-4.1",
+            "gpt-4.1-mini",
+            "gpt-4.1-nano",
+            "o3",
+            "o3-mini",
+            "o3-pro",
+            "o1",
+            "o1-pro",
+            "gpt-4o",
+            "gpt-4o-mini",
+        ],
     },
     anthropic: {
         agentName: "anthropicAgent",
         defaultModel: "claude-3-7-sonnet-20250219",
         max_tokens: 8192,
+        models: ["claude-opus-4-1-20250805", "claude-opus-4-20250514", "claude-sonnet-4-20250514", "claude-3-7-sonnet-20250219", "claude-3-haiku-20240307"],
     },
     gemini: {
         agentName: "geminiAgent",
-        defaultModel: "gemini-1.5-flash",
+        defaultModel: "gemini-2.5-flash",
         max_tokens: 8192,
+        models: ["gemini-2.5-pro", "gemini-2.5-flash", "gemini-2.5-flash-lite", "gemini-2.0-flash"],
     },
     groq: {
         agentName: "groqAgent",
-        defaultModel: "llama3-8b-8192",
+        defaultModel: "llama-3.1-8b-instant",
         max_tokens: 4096,
+        models: ["llama-3.1-8b-instant", "llama-3.3-70b-versatile", "deepseek-r1-distill-llama-70b", "openai/gpt-oss-120b", "openai/gpt-oss-20b"],
     },
     mock: {
         agentName: "mediaMockAgent",
         defaultModel: "mock",
         max_tokens: 4096,
+        models: ["mock"],
     },
 };
 export const defaultProviders = {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mulmocast",
-  "version": "1.2.3",
+  "version": "1.2.5",
   "description": "",
   "type": "module",
   "main": "lib/index.node.js",
@@ -67,27 +67,27 @@
   "dependencies": {
     "@google-cloud/text-to-speech": "^6.2.0",
     "@google/genai": "^1.13.0",
-    "@graphai/anthropic_agent": "^2.0.5",
+    "@graphai/anthropic_agent": "^2.0.11",
     "@graphai/browserless_agent": "^2.0.1",
-    "@graphai/gemini_agent": "^2.0.0",
-    "@graphai/groq_agent": "^2.0.0",
+    "@graphai/gemini_agent": "^2.0.1",
+    "@graphai/groq_agent": "^2.0.2",
     "@graphai/input_agents": "^1.0.2",
-    "@graphai/openai_agent": "^2.0.3",
+    "@graphai/openai_agent": "^2.0.5",
     "@graphai/stream_agent_filter": "^2.0.2",
-    "@graphai/vanilla": "^2.0.6",
-    "@graphai/vanilla_node_agents": "^2.0.1",
+    "@graphai/vanilla": "^2.0.12",
+    "@graphai/vanilla_node_agents": "^2.0.4",
     "@inquirer/input": "^4.2.1",
     "@inquirer/select": "^4.3.1",
     "@modelcontextprotocol/sdk": "^1.15.1",
     "@tavily/core": "^0.5.9",
-    "canvas": "^3.1.2",
+    "canvas": "^3.2.0",
     "clipboardy": "^4.0.0",
     "dotenv": "^17.2.1",
     "fluent-ffmpeg": "^2.1.3",
-    "graphai": "^2.0.13",
-    "marked": "^16.1.2",
+    "graphai": "^2.0.14",
+    "marked": "^16.2.0",
     "ora": "^8.2.0",
-    "puppeteer": "^24.16.0",
+    "puppeteer": "^24.16.2",
     "replicate": "^1.0.1",
     "yaml": "^2.8.1",
     "yargs": "^18.0.0",
@@ -97,7 +97,7 @@
   "devDependencies": {
     "@anatine/zod-mock": "^3.14.0",
     "@faker-js/faker": "^9.9.0",
-    "@receptron/test_utils": "^2.0.0",
+    "@receptron/test_utils": "^2.0.3",
     "@types/fluent-ffmpeg": "^2.1.26",
     "@types/yargs": "^17.0.33",
     "eslint": "^9.33.0",
@@ -106,9 +106,9 @@
     "eslint-plugin-sonarjs": "^3.0.4",
     "prettier": "^3.6.2",
     "ts-node": "^10.9.2",
-    "tsx": "^4.20.3",
+    "tsx": "^4.20.4",
     "typescript": "^5.9.2",
-    "typescript-eslint": "^8.39.0"
+    "typescript-eslint": "^8.40.0"
   },
   "engines": {
     "node": ">=18.0.0"

package/scripts/test/test_hello_caption.json~

@@ -1,21 +0,0 @@
-{
-  "$mulmocast": {
-    "version": "1.1"
-  },
-  "lang": "en",
-  "captionParams": {
-    "lang": "en"
-  },
-  "beats": [
-    {
-      "text": "Hello World",
-      "image": {
-        "type": "textSlide",
-        "slide": {
-          "title": "Hello World",
-          "bullets": ["Hello", "World"]
-        }
-      }
-    }
-  ]
-}

package/scripts/test/test_hello_image.json~

@@ -1,18 +0,0 @@
-{
-  "$mulmocast": {
-    "version": "1.1"
-  },
-  "lang": "en",
-  "beats": [
-    {
-      "text": "Hello World",
-      "image": {
-        "type": "textSlide",
-        "slide": {
-          "title": "Hello World",
-          "bullets": ["Hello", "World"]
-        }
-      }
-    }
-  ]
-}