mulmocast 2.1.7 → 2.1.9

package/lib/actions/audio.js

@@ -38,6 +38,8 @@ export const getBeatAudioPathOrUrl = (text, context, beat, lang) => {
         provider,
         model ?? "",
         speechOptions?.decoration ?? "",
+        speechOptions?.stability ?? "",
+        speechOptions?.similarity_boost ?? "",
     ].join(":");
     GraphAILogger.log(`getBeatAudioPathOrUrl [${hash_string}]`);
     const audioFileName = `${context.studio.filename}_${text2hash(hash_string)}`;
@@ -117,6 +119,8 @@ const graph_tts = {
                     speed: ":preprocessor.speechOptions.speed",
                     instructions: ":preprocessor.speechOptions.instruction",
                     decoration: ":preprocessor.speechOptions.decoration",
+                    stability: ":preprocessor.speechOptions.stability",
+                    similarityBoost: ":preprocessor.speechOptions.similarity_boost",
                     model: ":preprocessor.model",
                 },
             },

package/lib/actions/bundle.js

@@ -6,6 +6,7 @@ import { mkdir } from "../utils/file.js";
 import { ZipBuilder } from "../utils/zip.js";
 import { bundleTargetLang } from "../utils/const.js";
 import { createSilentAudio } from "../utils/ffmpeg_utils.js";
+import { silentMp3 } from "../utils/context.js";
 const downloadFile = async (url, destPath) => {
     const response = await fetch(url);
     if (!response.ok) {
@@ -53,7 +54,7 @@ const imageSourceMappings = [
 ];
 export const mulmoViewerBundle = async (context) => {
     const isZip = true;
-    const dir = path.resolve(context.fileDirs.fileName);
+    const dir = context.fileDirs.outDirPath;
     mkdir(dir);
     const zipper = new ZipBuilder(path.resolve(dir, zipFileName));
     // text
@@ -67,18 +68,24 @@ export const mulmoViewerBundle = async (context) => {
     // audio
     for (const lang of bundleTargetLang) {
         const audios = listLocalizedAudioPaths({ ...context, lang });
-        audios.forEach((audio, index) => {
+        await Promise.all(audios.map(async (audio, index) => {
             if (audio) {
                 const fileName = path.basename(audio ?? "");
                 if (resultJson[index] && resultJson[index].audioSources) {
                     resultJson[index].audioSources[lang] = fileName;
                 }
-                if (fs.existsSync(audio)) {
+                if (fileName === "silent300.mp3") {
+                    // Download from GitHub URL
+                    const destPath = path.resolve(dir, fileName);
+                    await downloadFile(silentMp3, destPath);
+                    zipper.addFile(destPath, fileName);
+                }
+                else if (fs.existsSync(audio)) {
                     fs.copyFileSync(audio, path.resolve(dir, fileName));
                     zipper.addFile(audio, fileName);
                 }
             }
-        });
+        }));
     }
     // image, movie
     context.studio.beats.forEach((image, index) => {

package/lib/agents/image_openai_agent.js

@@ -2,7 +2,7 @@ import fs from "fs";
 import path from "path";
 import { GraphAILogger } from "graphai";
 import OpenAI, { toFile, AuthenticationError, RateLimitError, APIError } from "openai";
-import { provider2ImageAgent } from "../utils/provider2agent.js";
+import { provider2ImageAgent, gptImages } from "../utils/provider2agent.js";
 import { apiKeyMissingError, agentGenerationError, openAIAgentGenerationError, agentIncorrectAPIKeyError, agentAPIRateLimitError, agentInvalidResponseError, imageAction, imageFileTarget, } from "../utils/error_cause.js";
 // https://platform.openai.com/docs/guides/image-generation
 export const imageOpenaiAgent = async ({ namedInputs, params, config, }) => {
@@ -17,7 +17,7 @@ export const imageOpenaiAgent = async ({ namedInputs, params, config, }) => {
     const model = params.model ?? provider2ImageAgent["openai"].defaultModel;
     const openai = new OpenAI({ apiKey, baseURL });
     const size = (() => {
-        if (model === "gpt-image-1" || model === "gpt-image-1-mini") {
+        if (gptImages.includes(model)) {
             if (canvasSize.width > canvasSize.height) {
                 return "1536x1024";
             }
@@ -46,7 +46,7 @@ export const imageOpenaiAgent = async ({ namedInputs, params, config, }) => {
         n: 1,
         size,
     };
-    if (model === "gpt-image-1" || model === "gpt-image-1-mini") {
+    if (gptImages.includes(model)) {
         imageOptions.moderation = moderation || "auto";
         imageOptions.background = "opaque";
         if (quality) {

package/lib/agents/tts_elevenlabs_agent.js

@@ -3,7 +3,7 @@ import { provider2TTSAgent } from "../utils/provider2agent.js";
 import { apiKeyMissingError, agentVoiceLimitReachedError, agentIncorrectAPIKeyError, agentGenerationError, audioAction, audioFileTarget, } from "../utils/error_cause.js";
 export const ttsElevenlabsAgent = async ({ namedInputs, params, config, }) => {
     const { text } = namedInputs;
-    const { voice, model, stability, similarityBoost, suppressError } = params;
+    const { voice, model, stability, similarityBoost, speed, suppressError } = params;
     const apiKey = config?.apiKey;
     if (!apiKey) {
         throw new Error("ElevenLabs API key is required (ELEVENLABS_API_KEY)", {
@@ -21,6 +21,7 @@ export const ttsElevenlabsAgent = async ({ namedInputs, params, config, }) => {
         voice_settings: {
             stability: stability ?? 0.5,
             similarity_boost: similarityBoost ?? 0.75,
+            speed: speed ?? 1.0,
         },
     };
     GraphAILogger.log("ElevenLabs TTS options", requestBody);

package/lib/types/agent.d.ts

@@ -127,4 +127,5 @@ export type ElevenlabsTTSAgentParams = TTSAgentParams & {
     model: string;
     stability: number;
     similarityBoost: number;
+    speed: number;
 };

package/lib/types/schema.d.ts

@@ -21,6 +21,8 @@ export declare const speechOptionsSchema: z.ZodObject<{
     speed: z.ZodOptional<z.ZodNumber>;
     instruction: z.ZodOptional<z.ZodString>;
     decoration: z.ZodOptional<z.ZodString>;
+    stability: z.ZodOptional<z.ZodNumber>;
+    similarity_boost: z.ZodOptional<z.ZodNumber>;
 }, z.core.$strict>;
 export declare const defaultSpeaker = "Presenter";
 export declare const text2SpeechProviderSchema: z.ZodDefault<z.ZodEnum<{
@@ -34,6 +36,8 @@ export declare const speakerDataSchema: z.ZodObject<{
         speed: z.ZodOptional<z.ZodNumber>;
         instruction: z.ZodOptional<z.ZodString>;
         decoration: z.ZodOptional<z.ZodString>;
+        stability: z.ZodOptional<z.ZodNumber>;
+        similarity_boost: z.ZodOptional<z.ZodNumber>;
     }, z.core.$strict>>;
     provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
         [x: string]: string;
@@ -48,6 +52,8 @@ export declare const speakerSchema: z.ZodObject<{
         speed: z.ZodOptional<z.ZodNumber>;
         instruction: z.ZodOptional<z.ZodString>;
         decoration: z.ZodOptional<z.ZodString>;
+        stability: z.ZodOptional<z.ZodNumber>;
+        similarity_boost: z.ZodOptional<z.ZodNumber>;
     }, z.core.$strict>>;
     provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
         [x: string]: string;
@@ -61,6 +67,8 @@ export declare const speakerSchema: z.ZodObject<{
             speed: z.ZodOptional<z.ZodNumber>;
             instruction: z.ZodOptional<z.ZodString>;
             decoration: z.ZodOptional<z.ZodString>;
+            stability: z.ZodOptional<z.ZodNumber>;
+            similarity_boost: z.ZodOptional<z.ZodNumber>;
         }, z.core.$strict>>;
         provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
             [x: string]: string;
@@ -76,6 +84,8 @@ export declare const speakerDictionarySchema: z.ZodRecord<z.ZodString, z.ZodObje
         speed: z.ZodOptional<z.ZodNumber>;
         instruction: z.ZodOptional<z.ZodString>;
         decoration: z.ZodOptional<z.ZodString>;
+        stability: z.ZodOptional<z.ZodNumber>;
+        similarity_boost: z.ZodOptional<z.ZodNumber>;
     }, z.core.$strict>>;
     provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
         [x: string]: string;
@@ -89,6 +99,8 @@ export declare const speakerDictionarySchema: z.ZodRecord<z.ZodString, z.ZodObje
             speed: z.ZodOptional<z.ZodNumber>;
             instruction: z.ZodOptional<z.ZodString>;
             decoration: z.ZodOptional<z.ZodString>;
+            stability: z.ZodOptional<z.ZodNumber>;
+            similarity_boost: z.ZodOptional<z.ZodNumber>;
         }, z.core.$strict>>;
         provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
             [x: string]: string;
@@ -105,6 +117,8 @@ export declare const mulmoSpeechParamsSchema: z.ZodDefault<z.ZodObject<{
             speed: z.ZodOptional<z.ZodNumber>;
             instruction: z.ZodOptional<z.ZodString>;
             decoration: z.ZodOptional<z.ZodString>;
+            stability: z.ZodOptional<z.ZodNumber>;
+            similarity_boost: z.ZodOptional<z.ZodNumber>;
         }, z.core.$strict>>;
         provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
             [x: string]: string;
@@ -118,6 +132,8 @@ export declare const mulmoSpeechParamsSchema: z.ZodDefault<z.ZodObject<{
                 speed: z.ZodOptional<z.ZodNumber>;
                 instruction: z.ZodOptional<z.ZodString>;
                 decoration: z.ZodOptional<z.ZodString>;
+                stability: z.ZodOptional<z.ZodNumber>;
+                similarity_boost: z.ZodOptional<z.ZodNumber>;
             }, z.core.$strict>>;
             provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                 [x: string]: string;
@@ -1105,6 +1121,8 @@ export declare const mulmoBeatSchema: z.ZodObject<{
         speed: z.ZodOptional<z.ZodNumber>;
         instruction: z.ZodOptional<z.ZodString>;
         decoration: z.ZodOptional<z.ZodString>;
+        stability: z.ZodOptional<z.ZodNumber>;
+        similarity_boost: z.ZodOptional<z.ZodNumber>;
     }, z.core.$strict>>;
     textSlideParams: z.ZodOptional<z.ZodObject<{
         cssStyles: z.ZodUnion<readonly [z.ZodString, z.ZodArray<z.ZodString>]>;
@@ -1167,6 +1185,8 @@ export declare const mulmoPresentationStyleSchema: z.ZodObject<{
                 speed: z.ZodOptional<z.ZodNumber>;
                 instruction: z.ZodOptional<z.ZodString>;
                 decoration: z.ZodOptional<z.ZodString>;
+                stability: z.ZodOptional<z.ZodNumber>;
+                similarity_boost: z.ZodOptional<z.ZodNumber>;
             }, z.core.$strict>>;
             provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                 [x: string]: string;
@@ -1180,6 +1200,8 @@ export declare const mulmoPresentationStyleSchema: z.ZodObject<{
                     speed: z.ZodOptional<z.ZodNumber>;
                     instruction: z.ZodOptional<z.ZodString>;
                     decoration: z.ZodOptional<z.ZodString>;
+                    stability: z.ZodOptional<z.ZodNumber>;
+                    similarity_boost: z.ZodOptional<z.ZodNumber>;
                 }, z.core.$strict>>;
                 provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                     [x: string]: string;
@@ -1514,6 +1536,8 @@ export declare const mulmoScriptSchema: z.ZodObject<{
                 speed: z.ZodOptional<z.ZodNumber>;
                 instruction: z.ZodOptional<z.ZodString>;
                 decoration: z.ZodOptional<z.ZodString>;
+                stability: z.ZodOptional<z.ZodNumber>;
+                similarity_boost: z.ZodOptional<z.ZodNumber>;
             }, z.core.$strict>>;
             provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                 [x: string]: string;
@@ -1527,6 +1551,8 @@ export declare const mulmoScriptSchema: z.ZodObject<{
                     speed: z.ZodOptional<z.ZodNumber>;
                     instruction: z.ZodOptional<z.ZodString>;
                     decoration: z.ZodOptional<z.ZodString>;
+                    stability: z.ZodOptional<z.ZodNumber>;
+                    similarity_boost: z.ZodOptional<z.ZodNumber>;
                 }, z.core.$strict>>;
                 provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                     [x: string]: string;
@@ -2222,6 +2248,8 @@ export declare const mulmoScriptSchema: z.ZodObject<{
             speed: z.ZodOptional<z.ZodNumber>;
             instruction: z.ZodOptional<z.ZodString>;
             decoration: z.ZodOptional<z.ZodString>;
+            stability: z.ZodOptional<z.ZodNumber>;
+            similarity_boost: z.ZodOptional<z.ZodNumber>;
         }, z.core.$strict>>;
         textSlideParams: z.ZodOptional<z.ZodObject<{
             cssStyles: z.ZodUnion<readonly [z.ZodString, z.ZodArray<z.ZodString>]>;
@@ -2354,6 +2382,8 @@ export declare const mulmoStudioSchema: z.ZodObject<{
                     speed: z.ZodOptional<z.ZodNumber>;
                     instruction: z.ZodOptional<z.ZodString>;
                     decoration: z.ZodOptional<z.ZodString>;
+                    stability: z.ZodOptional<z.ZodNumber>;
+                    similarity_boost: z.ZodOptional<z.ZodNumber>;
                 }, z.core.$strict>>;
                 provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                     [x: string]: string;
@@ -2367,6 +2397,8 @@ export declare const mulmoStudioSchema: z.ZodObject<{
                         speed: z.ZodOptional<z.ZodNumber>;
                         instruction: z.ZodOptional<z.ZodString>;
                         decoration: z.ZodOptional<z.ZodString>;
+                        stability: z.ZodOptional<z.ZodNumber>;
+                        similarity_boost: z.ZodOptional<z.ZodNumber>;
                     }, z.core.$strict>>;
                     provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                         [x: string]: string;
@@ -3062,6 +3094,8 @@ export declare const mulmoStudioSchema: z.ZodObject<{
                 speed: z.ZodOptional<z.ZodNumber>;
                 instruction: z.ZodOptional<z.ZodString>;
                 decoration: z.ZodOptional<z.ZodString>;
+                stability: z.ZodOptional<z.ZodNumber>;
+                similarity_boost: z.ZodOptional<z.ZodNumber>;
             }, z.core.$strict>>;
             textSlideParams: z.ZodOptional<z.ZodObject<{
                 cssStyles: z.ZodUnion<readonly [z.ZodString, z.ZodArray<z.ZodString>]>;
@@ -3130,6 +3164,8 @@ export declare const mulmoPromptTemplateSchema: z.ZodObject<{
                     speed: z.ZodOptional<z.ZodNumber>;
                     instruction: z.ZodOptional<z.ZodString>;
                     decoration: z.ZodOptional<z.ZodString>;
+                    stability: z.ZodOptional<z.ZodNumber>;
+                    similarity_boost: z.ZodOptional<z.ZodNumber>;
                 }, z.core.$strict>>;
                 provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                     [x: string]: string;
@@ -3143,6 +3179,8 @@ export declare const mulmoPromptTemplateSchema: z.ZodObject<{
                         speed: z.ZodOptional<z.ZodNumber>;
                         instruction: z.ZodOptional<z.ZodString>;
                         decoration: z.ZodOptional<z.ZodString>;
+                        stability: z.ZodOptional<z.ZodNumber>;
+                        similarity_boost: z.ZodOptional<z.ZodNumber>;
                     }, z.core.$strict>>;
                     provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                         [x: string]: string;
@@ -3471,6 +3509,8 @@ export declare const mulmoPromptTemplateFileSchema: z.ZodObject<{
                     speed: z.ZodOptional<z.ZodNumber>;
                     instruction: z.ZodOptional<z.ZodString>;
                     decoration: z.ZodOptional<z.ZodString>;
+                    stability: z.ZodOptional<z.ZodNumber>;
+                    similarity_boost: z.ZodOptional<z.ZodNumber>;
                 }, z.core.$strict>>;
                 provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                     [x: string]: string;
@@ -3484,6 +3524,8 @@ export declare const mulmoPromptTemplateFileSchema: z.ZodObject<{
                         speed: z.ZodOptional<z.ZodNumber>;
                         instruction: z.ZodOptional<z.ZodString>;
                         decoration: z.ZodOptional<z.ZodString>;
+                        stability: z.ZodOptional<z.ZodNumber>;
+                        similarity_boost: z.ZodOptional<z.ZodNumber>;
                     }, z.core.$strict>>;
                     provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                         [x: string]: string;

package/lib/types/schema.js

@@ -21,9 +21,11 @@ export const localizedTextSchema = z
 export const multiLingualTextsSchema = z.record(langSchema, localizedTextSchema);
 export const speechOptionsSchema = z
     .object({
-    speed: z.number().optional(), // default: 1.0 for google and niji voice
+    speed: z.number().optional(), // default: 1.0 for google and niji voice, elevenLabs
     instruction: z.string().optional(), // for tts openai
     decoration: z.string().optional(), // for kotodama. default: neutral
+    stability: z.number().optional(), // for elevenLabs
+    similarity_boost: z.number().optional(), // for elevenLabs
 })
     .strict();
 const speakerIdSchema = z.string();

package/lib/utils/context.d.ts

@@ -1,4 +1,5 @@
 import type { MulmoStudioBeat, MulmoScript, MulmoPresentationStyle, MulmoStudioMultiLingual, FileObject } from "../types/type.js";
+export declare const silentMp3 = "https://github.com/receptron/mulmocast-cli/raw/refs/heads/main/assets/audio/silent300.mp3";
 export declare const createStudioData: (_mulmoScript: MulmoScript, fileName: string, videoCaptionLang?: string, presentationStyle?: MulmoPresentationStyle | null) => {
     script: {
         $mulmocast: {
@@ -18,6 +19,8 @@ export declare const createStudioData: (_mulmoScript: MulmoScript, fileName: str
                     speed?: number | undefined;
                     instruction?: string | undefined;
                     decoration?: string | undefined;
+                    stability?: number | undefined;
+                    similarity_boost?: number | undefined;
                 } | undefined;
                 provider?: string | undefined;
                 model?: string | undefined;
@@ -29,6 +32,8 @@ export declare const createStudioData: (_mulmoScript: MulmoScript, fileName: str
                         speed?: number | undefined;
                         instruction?: string | undefined;
                         decoration?: string | undefined;
+                        stability?: number | undefined;
+                        similarity_boost?: number | undefined;
                     } | undefined;
                     provider?: string | undefined;
                     model?: string | undefined;
@@ -541,6 +546,8 @@ export declare const createStudioData: (_mulmoScript: MulmoScript, fileName: str
                 speed?: number | undefined;
                 instruction?: string | undefined;
                 decoration?: string | undefined;
+                stability?: number | undefined;
+                similarity_boost?: number | undefined;
             } | undefined;
             textSlideParams?: {
                 cssStyles: string | string[];
@@ -632,6 +639,8 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                         speed?: number | undefined;
                         instruction?: string | undefined;
                         decoration?: string | undefined;
+                        stability?: number | undefined;
+                        similarity_boost?: number | undefined;
                     } | undefined;
                     provider?: string | undefined;
                     model?: string | undefined;
@@ -643,6 +652,8 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                             speed?: number | undefined;
                             instruction?: string | undefined;
                             decoration?: string | undefined;
+                            stability?: number | undefined;
+                            similarity_boost?: number | undefined;
                         } | undefined;
                         provider?: string | undefined;
                         model?: string | undefined;
@@ -1155,6 +1166,8 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                     speed?: number | undefined;
                     instruction?: string | undefined;
                     decoration?: string | undefined;
+                    stability?: number | undefined;
+                    similarity_boost?: number | undefined;
                 } | undefined;
                 textSlideParams?: {
                     cssStyles: string | string[];
@@ -1253,6 +1266,8 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                     speed?: number | undefined;
                     instruction?: string | undefined;
                     decoration?: string | undefined;
+                    stability?: number | undefined;
+                    similarity_boost?: number | undefined;
                 } | undefined;
                 provider?: string | undefined;
                 model?: string | undefined;
@@ -1264,6 +1279,8 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                         speed?: number | undefined;
                         instruction?: string | undefined;
                         decoration?: string | undefined;
+                        stability?: number | undefined;
+                        similarity_boost?: number | undefined;
                     } | undefined;
                     provider?: string | undefined;
                     model?: string | undefined;

package/lib/utils/context.js

@@ -3,6 +3,7 @@ import { readMulmoScriptFile, fetchMulmoScriptFile, isFile } from "./file.js";
 import { beatId, multiLingualObjectToArray } from "./utils.js";
 import { mulmoStudioSchema, mulmoCaptionParamsSchema, mulmoPresentationStyleSchema } from "../types/schema.js";
 import { MulmoPresentationStyleMethods, MulmoScriptMethods, MulmoStudioMultiLingualMethod } from "../methods/index.js";
+export const silentMp3 = "https://github.com/receptron/mulmocast-cli/raw/refs/heads/main/assets/audio/silent300.mp3";
 const mulmoCredit = (speaker) => {
     return {
         id: "mulmo_credit",
@@ -19,7 +20,7 @@ const mulmoCredit = (speaker) => {
             type: "audio",
             source: {
                 kind: "url",
-                url: "https://github.com/receptron/mulmocast-cli/raw/refs/heads/main/assets/audio/silent300.mp3",
+                url: silentMp3,
             },
         },
     };

package/lib/utils/provider2agent.d.ts

@@ -46,6 +46,7 @@ export declare const provider2TTSAgent: {
         models: string[];
     };
 };
+export declare const gptImages: string[];
 export declare const provider2ImageAgent: {
     openai: {
         agentName: string;

package/lib/utils/provider2agent.js

@@ -49,11 +49,12 @@ export const provider2TTSAgent = {
         models: ["mock-model"],
     },
 };
+export const gptImages = ["gpt-image-1.5", "gpt-image-1", "gpt-image-1-mini"];
 export const provider2ImageAgent = {
     openai: {
         agentName: "imageOpenaiAgent",
         defaultModel: "gpt-image-1",
-        models: ["dall-e-3", "gpt-image-1"],
+        models: ["dall-e-3", ...gptImages],
         keyName: "OPENAI_API_KEY",
         baseURLKeyName: "OPENAI_BASE_URL",
     },

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mulmocast",
-  "version": "2.1.7",
+  "version": "2.1.9",
   "description": "",
   "type": "module",
   "main": "lib/index.node.js",
@@ -74,7 +74,7 @@
   "homepage": "https://github.com/receptron/mulmocast-cli#readme",
   "dependencies": {
     "@google-cloud/text-to-speech": "^6.4.0",
-    "@google/genai": "^1.32.0",
+    "@google/genai": "^1.34.0",
     "@graphai/anthropic_agent": "^2.0.12",
     "@graphai/browserless_agent": "^2.0.1",
     "@graphai/gemini_agent": "^2.0.1",
@@ -86,7 +86,7 @@
     "@graphai/vanilla_node_agents": "^2.0.4",
     "@inquirer/input": "^4.3.0",
     "@inquirer/select": "^4.4.1",
-    "@modelcontextprotocol/sdk": "^1.22.0",
+    "@modelcontextprotocol/sdk": "^1.25.1",
     "@mozilla/readability": "^0.6.0",
     "@tavily/core": "^0.5.11",
     "archiver": "^7.0.1",
@@ -98,11 +98,11 @@
     "marked": "^17.0.1",
     "mulmocast-vision": "^1.0.8",
     "ora": "^9.0.0",
-    "puppeteer": "^24.32.1",
+    "puppeteer": "^24.33.0",
     "replicate": "^1.4.0",
     "yaml": "^2.8.2",
     "yargs": "^18.0.0",
-    "zod": "^4.1.13"
+    "zod": "^4.2.1"
   },
   "devDependencies": {
     "@receptron/test_utils": "^2.0.3",
@@ -110,14 +110,14 @@
     "@types/fluent-ffmpeg": "^2.1.28",
     "@types/jsdom": "^27.0.0",
     "@types/yargs": "^17.0.35",
-    "eslint": "^9.39.1",
+    "eslint": "^9.39.2",
     "eslint-config-prettier": "^10.1.8",
     "eslint-plugin-prettier": "^5.5.4",
     "eslint-plugin-sonarjs": "^3.0.5",
     "prettier": "^3.7.4",
     "tsx": "^4.21.0",
     "typescript": "^5.9.3",
-    "typescript-eslint": "^8.49.0"
+    "typescript-eslint": "^8.50.0"
   },
   "engines": {
     "node": ">=20.0.0"

package/scripts/test/test_all_image.json

@@ -18,6 +18,15 @@
         "model": "gpt-image-1"
       }
     },
+    {
+      "speaker": "Presenter",
+      "text": "こんにちは、テストです。",
+      "imagePrompt": "美しい日本庭園",
+      "imageParams": {
+        "provider": "openai",
+        "model": "gpt-image-1.5"
+      }
+    },
     {
       "speaker": "Presenter",
       "text": "こんにちは、テストです。",

package/scripts/test/test_elevenlabs_option.json

@@ -0,0 +1,30 @@
+{
+  "$mulmocast": {
+    "version": "1.0",
+    "credit": "closing"
+  },
+  "speechParams": {
+    "speakers": {
+      "otan": {
+        "provider": "elevenlabs",
+        "voiceId": "21m00Tcm4TlvDq8ikWAM",
+        "speechOptions": {
+          "speed": 0.8,
+          "stability": 0.6,
+          "similarity_boost": 0.8
+        }
+      }
+    }
+  },
+  "beats": [
+    {
+      "id": "otan",
+      "speaker": "otan",
+      "text": "こんにちは, 良い天気です",
+      "image": {
+        "type": "markdown",
+        "markdown": ["# Voice is otan"]
+      }
+    }
+  ]
+}

package/lib/agents/test.d.ts

@@ -1 +0,0 @@
-import "dotenv/config";

package/lib/agents/test.js

@@ -1,12 +0,0 @@
-import "dotenv/config";
-import { ttsKotodamaAgent } from "./tts_kotodama_agent.js";
-const kotodamaApiKey = process.env.KOTODAMA_API_KEY ?? "";
-const main = async () => {
-    const result = await ttsKotodamaAgent({
-        namedInputs: { text: "こんにちは" },
-        params: { voice: "Atla", decoration: "neutral", suppressError: false },
-        config: { apiKey: kotodamaApiKey },
-    });
-    console.log("Result:", result);
-};
-main();

package/lib/utils/agent_error_handler.d.ts

@@ -1,24 +0,0 @@
-/**
- * Wraps agent execution with standardized error handling
- *
- * @param agentName - Name of the agent for error reporting
- * @param action - Action being performed (e.g., "audio", "image", "movie")
- * @param target - Target of the action (e.g., "audioFile", "imageFile")
- * @param suppressError - If true, returns error object instead of throwing
- * @param fn - Async function to execute
- * @returns Result of fn or error object if suppressError is true
- */
-export declare const wrapAgentError: <T>(agentName: string, action: string, target: string, suppressError: boolean | undefined, fn: () => Promise<T>) => Promise<T | {
-    error: unknown;
-}>;
-/**
- * Validates that an API key is present, throwing an error if missing
- *
- * @param apiKey - API key to validate
- * @param agentName - Name of the agent for error reporting
- * @param action - Action being performed
- * @param envVarName - Environment variable name for the API key
- * @returns The validated API key
- * @throws Error if API key is missing
- */
-export declare const requireApiKey: (apiKey: string | undefined, agentName: string, action: string, envVarName: string) => string;

package/lib/utils/agent_error_handler.js

@@ -1,59 +0,0 @@
-import { GraphAILogger } from "graphai";
-import { AuthenticationError, RateLimitError } from "openai";
-import { agentIncorrectAPIKeyError, agentAPIRateLimitError, agentGenerationError, hasCause, } from "./error_cause.js";
-/**
- * Wraps agent execution with standardized error handling
- *
- * @param agentName - Name of the agent for error reporting
- * @param action - Action being performed (e.g., "audio", "image", "movie")
- * @param target - Target of the action (e.g., "audioFile", "imageFile")
- * @param suppressError - If true, returns error object instead of throwing
- * @param fn - Async function to execute
- * @returns Result of fn or error object if suppressError is true
- */
-export const wrapAgentError = async (agentName, action, target, suppressError, fn) => {
-    try {
-        return await fn();
-    }
-    catch (error) {
-        if (suppressError) {
-            return { error };
-        }
-        GraphAILogger.info(error);
-        if (error instanceof AuthenticationError) {
-            throw new Error(`Failed: 401 Incorrect API key provided`, {
-                cause: agentIncorrectAPIKeyError(agentName, action, target),
-            });
-        }
-        if (error instanceof RateLimitError) {
-            throw new Error("You exceeded your current quota, please check your plan and billing details", {
-                cause: agentAPIRateLimitError(agentName, action, target),
-            });
-        }
-        if (hasCause(error)) {
-            throw error;
-        }
-        throw new Error(`Failed with ${agentName}`, {
-            cause: agentGenerationError(agentName, action, target),
-        });
-    }
-};
-/**
- * Validates that an API key is present, throwing an error if missing
- *
- * @param apiKey - API key to validate
- * @param agentName - Name of the agent for error reporting
- * @param action - Action being performed
- * @param envVarName - Environment variable name for the API key
- * @returns The validated API key
- * @throws Error if API key is missing
- */
-export const requireApiKey = (apiKey, agentName, action, envVarName) => {
-    if (!apiKey) {
-        const { apiKeyMissingError } = require("./error_cause.js");
-        throw new Error(`${envVarName} is required`, {
-            cause: apiKeyMissingError(agentName, action, envVarName),
-        });
-    }
-    return apiKey;
-};

package/lib/utils/agent_info_factory.d.ts

@@ -1,12 +0,0 @@
-import type { AgentFunction, AgentFunctionInfo } from "graphai";
-/**
- * Creates standardized AgentFunctionInfo metadata for agents
- *
- * @param name - Agent name (e.g., "ttsOpenaiAgent")
- * @param agent - Agent function implementation
- * @param category - Agent categories (e.g., ["tts"], ["image"])
- * @param description - Human-readable description
- * @param environmentVariables - Required environment variables (e.g., ["OPENAI_API_KEY"])
- * @returns AgentFunctionInfo with standardized metadata
- */
-export declare const createAgentInfo: <P = unknown, R = unknown, I = unknown, C = unknown>(name: string, agent: AgentFunction<P, R, I, C>, category: string[], description: string, environmentVariables?: string[]) => AgentFunctionInfo;

package/lib/utils/agent_info_factory.js

@@ -1,23 +0,0 @@
-import { PROJECT_METADATA } from "./const.js";
-/**
- * Creates standardized AgentFunctionInfo metadata for agents
- *
- * @param name - Agent name (e.g., "ttsOpenaiAgent")
- * @param agent - Agent function implementation
- * @param category - Agent categories (e.g., ["tts"], ["image"])
- * @param description - Human-readable description
- * @param environmentVariables - Required environment variables (e.g., ["OPENAI_API_KEY"])
- * @returns AgentFunctionInfo with standardized metadata
- */
-export const createAgentInfo = (name, agent, category, description, environmentVariables = []) => ({
-    name,
-    agent,
-    mock: agent,
-    samples: [],
-    description,
-    category,
-    author: PROJECT_METADATA.author,
-    repository: PROJECT_METADATA.repository,
-    license: PROJECT_METADATA.license,
-    environmentVariables,
-});

package/scripts/test/test_transition2.json~

@@ -1,62 +0,0 @@
-{
-  "$mulmocast": {
-    "version": "1.1"
-  },
-  "lang": "en",
-  "title": "Transition Test",
-  "movieParams": {
-    "transition": {
-      "type": "fade",
-      "duration": 1.0
-    }
-  },
-  "beats": [
-    {
-      "speaker": "Presenter",
-      "text": "",
-      "image": {
-        "type": "image",
-        "source": {
-          "kind": "path",
-          "path": "../../assets/images/mulmocast_credit.png"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "duration": 3.0,
-      "image": {
-        "type": "textSlide",
-        "slide": {
-          "title": "3.0 second with no Audio"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "duration": 3.0,
-      "movieParams": {
-        "transition": {
-          "type": "slideout_left",
-          "duration": 1.0
-        }
-      },
-      "image": {
-        "type": "textSlide",
-        "slide": {
-          "title": "3.0 second with no Audio"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/test/pingpong.mov"
-        }
-      }
-    }
-  ]
-}

package/scripts/test/test_transition3.json~

@@ -1,76 +0,0 @@
-{
-  "$mulmocast": {
-    "version": "1.1"
-  },
-  "lang": "en",
-  "title": "Transition Test",
-  "movieParams": {
-    "transition": {
-      "type": "fade",
-      "duration": 1.0
-    }
-  },
-  "beats": [
-    {
-      "speaker": "Presenter",
-      "text": "",
-      "image": {
-        "type": "image",
-        "source": {
-          "kind": "path",
-          "path": "../../assets/images/mulmocast_credit.png"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "image": {
-        "type": "textSlide",
-        "slide": {
-          "title": "1"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "movieParams": {
-        "transition": {
-          "type": "slideout_left",
-          "duration": 1.0
-        }
-      },
-      "image": {
-        "type": "textSlide",
-        "slide": {
-          "title": "2"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/tv.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-       "movieParams": {
-        "transition": {
-          "type": "slideout_left",
-          "duration": 1.0
-        }
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/test/pingpong.mov"
-        }
-      }
-    }
-  ]
-}

package/scripts/test/test_video_filters.json~

@@ -1,227 +0,0 @@
-{
-  "$mulmocast": {
-    "version": "1.1"
-  },
-  "lang": "en",
-  "title": "Video Filter Test",
-  "speechParams": {
-    "speakers": {
-      "Presenter": {
-        "voiceId": "shimmer",
-        "displayName": {
-          "en": "Presenter"
-        }
-      }
-    }
-  },
-  "beats": [
-    {
-      "speaker": "Presenter",
-      "text": "Original.",
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Mono filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "mono"
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Sepia filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "sepia"
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Brightness and contrast filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "brightness_contrast",
-            "brightness": 0.05,
-            "contrast": 1.3
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Box blur filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "blur",
-            "radius": 10,
-            "power": 1
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Gaussian blur filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "gblur",
-            "sigma": 30
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Glitch noise filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "glitch",
-            "intensity": 20,
-            "style": "noise"
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Glitch blend filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "glitch",
-            "intensity": 40,
-            "style": "blend"
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Grain filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "grain",
-            "intensity": 10
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Combined contrast and grain filters.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "brightness_contrast",
-            "brightness": 0,
-            "contrast": 1.2
-          },
-          {
-            "type": "grain",
-            "intensity": 6
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    },
-    {
-      "speaker": "Presenter",
-      "text": "Custom horizontal flip filter.",
-      "movieParams": {
-        "filters": [
-          {
-            "type": "custom",
-            "filter": "hflip"
-          }
-        ]
-      },
-      "image": {
-        "type": "movie",
-        "source": {
-          "kind": "url",
-          "url": "https://github.com/receptron/mulmocast-media/raw/refs/heads/main/movies/horse_galloping.mov"
-        }
-      }
-    }
-  ]
-}