@framers/agentos-ext-voice-synthesis 1.0.0

package/README.md

@@ -0,0 +1,28 @@
+# Voice Synthesis Extension for AgentOS
+
+Text-to-speech synthesis using ElevenLabs with 8 predefined voices.
+
+## Installation
+
+```bash
+npm install @framers/agentos-ext-voice-synthesis
+```
+
+## Configuration
+
+Set `ELEVENLABS_API_KEY` environment variable or pass via options.
+
+## Tool: text_to_speech
+
+**Input:**
+- `text` (string, required) — Text to convert (max 5000 chars)
+- `voice` (string, default: rachel) — Voice: rachel, domi, bella, antoni, josh, arnold, adam, sam
+- `model` (string, default: eleven_monolingual_v1) — ElevenLabs model
+- `stability` (number, 0-1, default: 0.5) — Voice stability
+- `similarity_boost` (number, 0-1, default: 0.75) — Voice similarity
+
+**Output:** Base64-encoded MP3 audio with duration estimate.
+
+## License
+
+MIT - Frame.dev

package/dist/index.d.ts

@@ -0,0 +1,27 @@
+/**
+ * Voice Synthesis Extension Pack — ElevenLabs TTS for agents.
+ */
+import { TextToSpeechTool } from './tools/textToSpeech.js';
+export interface VoiceSynthesisExtensionOptions {
+    elevenLabsApiKey?: string;
+    priority?: number;
+}
+export declare function createExtensionPack(context: any): {
+    name: string;
+    version: string;
+    descriptors: {
+        id: string;
+        kind: "tool";
+        priority: number;
+        payload: TextToSpeechTool;
+        requiredSecrets: {
+            id: string;
+        }[];
+    }[];
+    onActivate: () => Promise<any>;
+    onDeactivate: () => Promise<any>;
+};
+export { TextToSpeechTool };
+export type { TTSInput, TTSOutput } from './tools/textToSpeech.js';
+export default createExtensionPack;
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,EAAE,gBAAgB,EAAE,MAAM,yBAAyB,CAAC;AAE3D,MAAM,WAAW,8BAA8B;IAC7C,gBAAgB,CAAC,EAAE,MAAM,CAAC;IAC1B,QAAQ,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,wBAAgB,mBAAmB,CAAC,OAAO,EAAE,GAAG;;;;;;;;;;;;;;EAc/C;AAED,OAAO,EAAE,gBAAgB,EAAE,CAAC;AAC5B,YAAY,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,yBAAyB,CAAC;AACnE,eAAe,mBAAmB,CAAC"}

package/dist/index.js

@@ -0,0 +1,21 @@
+/**
+ * Voice Synthesis Extension Pack — ElevenLabs TTS for agents.
+ */
+import { TextToSpeechTool } from './tools/textToSpeech.js';
+export function createExtensionPack(context) {
+    const options = (context.options || {});
+    const apiKey = options.elevenLabsApiKey || context.getSecret?.('elevenlabs.apiKey') || process.env.ELEVENLABS_API_KEY;
+    const tool = new TextToSpeechTool(apiKey);
+    return {
+        name: '@framers/agentos-ext-voice-synthesis',
+        version: '1.0.0',
+        descriptors: [
+            { id: 'textToSpeech', kind: 'tool', priority: options.priority || 50, payload: tool, requiredSecrets: [{ id: 'elevenlabs.apiKey' }] },
+        ],
+        onActivate: async () => context.logger?.info('Voice Synthesis Extension activated'),
+        onDeactivate: async () => context.logger?.info('Voice Synthesis Extension deactivated'),
+    };
+}
+export { TextToSpeechTool };
+export default createExtensionPack;
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,EAAE,gBAAgB,EAAE,MAAM,yBAAyB,CAAC;AAO3D,MAAM,UAAU,mBAAmB,CAAC,OAAY;IAC9C,MAAM,OAAO,GAAG,CAAC,OAAO,CAAC,OAAO,IAAI,EAAE,CAAmC,CAAC;IAC1E,MAAM,MAAM,GAAG,OAAO,CAAC,gBAAgB,IAAI,OAAO,CAAC,SAAS,EAAE,CAAC,mBAAmB,CAAC,IAAI,OAAO,CAAC,GAAG,CAAC,kBAAkB,CAAC;IACtH,MAAM,IAAI,GAAG,IAAI,gBAAgB,CAAC,MAAM,CAAC,CAAC;IAE1C,OAAO;QACL,IAAI,EAAE,sCAAsC;QAC5C,OAAO,EAAE,OAAO;QAChB,WAAW,EAAE;YACX,EAAE,EAAE,EAAE,cAAc,EAAE,IAAI,EAAE,MAAe,EAAE,QAAQ,EAAE,OAAO,CAAC,QAAQ,IAAI,EAAE,EAAE,OAAO,EAAE,IAAI,EAAE,eAAe,EAAE,CAAC,EAAE,EAAE,EAAE,mBAAmB,EAAE,CAAC,EAAE;SAC/I;QACD,UAAU,EAAE,KAAK,IAAI,EAAE,CAAC,OAAO,CAAC,MAAM,EAAE,IAAI,CAAC,qCAAqC,CAAC;QACnF,YAAY,EAAE,KAAK,IAAI,EAAE,CAAC,OAAO,CAAC,MAAM,EAAE,IAAI,CAAC,uCAAuC,CAAC;KACxF,CAAC;AACJ,CAAC;AAED,OAAO,EAAE,gBAAgB,EAAE,CAAC;AAE5B,eAAe,mBAAmB,CAAC"}

package/dist/tools/textToSpeech.d.ts

@@ -0,0 +1,34 @@
+/**
+ * ElevenLabs TTS Tool — text-to-speech synthesis.
+ */
+import type { ITool, ToolExecutionContext, ToolExecutionResult, JSONSchemaObject } from '../../../../../../agentos/src/core/tools/ITool.js';
+export interface TTSInput {
+    text: string;
+    voice?: string;
+    model?: string;
+    stability?: number;
+    similarity_boost?: number;
+}
+export interface TTSOutput {
+    text: string;
+    voice: string;
+    model: string;
+    audioBase64: string;
+    contentType: string;
+    durationEstimateMs: number;
+}
+export declare class TextToSpeechTool implements ITool<TTSInput, TTSOutput> {
+    readonly id = "elevenlabs-tts-v1";
+    readonly name = "text_to_speech";
+    readonly displayName = "Text to Speech";
+    readonly description: string;
+    readonly category = "media";
+    readonly version = "1.0.0";
+    readonly hasSideEffects = false;
+    readonly inputSchema: JSONSchemaObject;
+    readonly requiredCapabilities: string[];
+    private apiKey;
+    constructor(apiKey?: string);
+    execute(args: TTSInput, _context: ToolExecutionContext): Promise<ToolExecutionResult<TTSOutput>>;
+}
+//# sourceMappingURL=textToSpeech.d.ts.map

package/dist/tools/textToSpeech.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"textToSpeech.d.ts","sourceRoot":"","sources":["../../src/tools/textToSpeech.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,KAAK,EAAE,KAAK,EAAE,oBAAoB,EAAE,mBAAmB,EAAE,gBAAgB,EAAE,MAAM,mDAAmD,CAAC;AAE5I,MAAM,WAAW,QAAQ;IACvB,IAAI,EAAE,MAAM,CAAC;IACb,KAAK,CAAC,EAAE,MAAM,CAAC;IACf,KAAK,CAAC,EAAE,MAAM,CAAC;IACf,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,gBAAgB,CAAC,EAAE,MAAM,CAAC;CAC3B;AAED,MAAM,WAAW,SAAS;IACxB,IAAI,EAAE,MAAM,CAAC;IACb,KAAK,EAAE,MAAM,CAAC;IACd,KAAK,EAAE,MAAM,CAAC;IACd,WAAW,EAAE,MAAM,CAAC;IACpB,WAAW,EAAE,MAAM,CAAC;IACpB,kBAAkB,EAAE,MAAM,CAAC;CAC5B;AAaD,qBAAa,gBAAiB,YAAW,KAAK,CAAC,QAAQ,EAAE,SAAS,CAAC;IACjE,QAAQ,CAAC,EAAE,uBAAuB;IAClC,QAAQ,CAAC,IAAI,oBAAoB;IACjC,QAAQ,CAAC,WAAW,oBAAoB;IACxC,QAAQ,CAAC,WAAW,SAE8C;IAClE,QAAQ,CAAC,QAAQ,WAAW;IAC5B,QAAQ,CAAC,OAAO,WAAW;IAC3B,QAAQ,CAAC,cAAc,SAAS;IAEhC,QAAQ,CAAC,WAAW,EAAE,gBAAgB,CAUpC;IAEF,QAAQ,CAAC,oBAAoB,WAAsB;IAEnD,OAAO,CAAC,MAAM,CAAS;gBAEX,MAAM,CAAC,EAAE,MAAM;IAIrB,OAAO,CAAC,IAAI,EAAE,QAAQ,EAAE,QAAQ,EAAE,oBAAoB,GAAG,OAAO,CAAC,mBAAmB,CAAC,SAAS,CAAC,CAAC;CAoCvG"}

package/dist/tools/textToSpeech.js

@@ -0,0 +1,73 @@
+/**
+ * ElevenLabs TTS Tool — text-to-speech synthesis.
+ */
+const VOICES = {
+    rachel: '21m00Tcm4TlvDq8ikWAM',
+    domi: 'AZnzlk1XvdvUeBnXmlld',
+    bella: 'EXAVITQu4vr4xnSDxMaL',
+    antoni: 'ErXwobaYiN019PkySvjV',
+    josh: 'TxGEqnHWrfWFTfGW9XjX',
+    arnold: 'VR6AewLTigWG4xSOukaG',
+    adam: 'pNInz6obpgDQGcFmaJgB',
+    sam: 'yoZ06aMxZJJ28mfd3POQ',
+};
+export class TextToSpeechTool {
+    id = 'elevenlabs-tts-v1';
+    name = 'text_to_speech';
+    displayName = 'Text to Speech';
+    description = 'Convert text to speech using ElevenLabs. Returns base64-encoded MP3 audio. ' +
+        'Voices: rachel, domi, bella, antoni, josh, arnold, adam, sam.';
+    category = 'media';
+    version = '1.0.0';
+    hasSideEffects = false;
+    inputSchema = {
+        type: 'object',
+        properties: {
+            text: { type: 'string', description: 'Text to convert. Max 5000 chars.' },
+            voice: { type: 'string', default: 'rachel', description: 'Voice name or ID.' },
+            model: { type: 'string', default: 'eleven_monolingual_v1' },
+            stability: { type: 'number', minimum: 0, maximum: 1, default: 0.5 },
+            similarity_boost: { type: 'number', minimum: 0, maximum: 1, default: 0.75 },
+        },
+        required: ['text'],
+    };
+    requiredCapabilities = ['capability:tts'];
+    apiKey;
+    constructor(apiKey) {
+        this.apiKey = apiKey || process.env.ELEVENLABS_API_KEY || '';
+    }
+    async execute(args, _context) {
+        if (!this.apiKey)
+            return { success: false, error: 'ELEVENLABS_API_KEY not configured.' };
+        const text = args.text.slice(0, 5000);
+        const voiceId = VOICES[(args.voice || 'rachel').toLowerCase()] || args.voice || VOICES.rachel;
+        const model = args.model || 'eleven_monolingual_v1';
+        try {
+            const response = await fetch(`https://api.elevenlabs.io/v1/text-to-speech/${voiceId}`, {
+                method: 'POST',
+                headers: { 'xi-api-key': this.apiKey, 'Content-Type': 'application/json', Accept: 'audio/mpeg' },
+                body: JSON.stringify({
+                    text,
+                    model_id: model,
+                    voice_settings: { stability: args.stability ?? 0.5, similarity_boost: args.similarity_boost ?? 0.75 },
+                }),
+            });
+            if (!response.ok) {
+                const err = await response.text();
+                return { success: false, error: `ElevenLabs error (${response.status}): ${err}` };
+            }
+            const buf = await response.arrayBuffer();
+            const audioBase64 = Buffer.from(buf).toString('base64');
+            const durationEstimateMs = Math.round((text.split(/\s+/).length / 150) * 60 * 1000);
+            return {
+                success: true,
+                output: { text, voice: args.voice || 'rachel', model, audioBase64, contentType: 'audio/mpeg', durationEstimateMs },
+                contentType: 'audio/mpeg',
+            };
+        }
+        catch (err) {
+            return { success: false, error: `TTS failed: ${err.message}` };
+        }
+    }
+}
+//# sourceMappingURL=textToSpeech.js.map

package/dist/tools/textToSpeech.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"textToSpeech.js","sourceRoot":"","sources":["../../src/tools/textToSpeech.ts"],"names":[],"mappings":"AAAA;;GAEG;AAqBH,MAAM,MAAM,GAA2B;IACrC,MAAM,EAAE,sBAAsB;IAC9B,IAAI,EAAE,sBAAsB;IAC5B,KAAK,EAAE,sBAAsB;IAC7B,MAAM,EAAE,sBAAsB;IAC9B,IAAI,EAAE,sBAAsB;IAC5B,MAAM,EAAE,sBAAsB;IAC9B,IAAI,EAAE,sBAAsB;IAC5B,GAAG,EAAE,sBAAsB;CAC5B,CAAC;AAEF,MAAM,OAAO,gBAAgB;IAClB,EAAE,GAAG,mBAAmB,CAAC;IACzB,IAAI,GAAG,gBAAgB,CAAC;IACxB,WAAW,GAAG,gBAAgB,CAAC;IAC/B,WAAW,GAClB,6EAA6E;QAC7E,+DAA+D,CAAC;IACzD,QAAQ,GAAG,OAAO,CAAC;IACnB,OAAO,GAAG,OAAO,CAAC;IAClB,cAAc,GAAG,KAAK,CAAC;IAEvB,WAAW,GAAqB;QACvC,IAAI,EAAE,QAAQ;QACd,UAAU,EAAE;YACV,IAAI,EAAE,EAAE,IAAI,EAAE,QAAQ,EAAE,WAAW,EAAE,kCAAkC,EAAE;YACzE,KAAK,EAAE,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,QAAQ,EAAE,WAAW,EAAE,mBAAmB,EAAE;YAC9E,KAAK,EAAE,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,uBAAuB,EAAE;YAC3D,SAAS,EAAE,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,CAAC,EAAE,OAAO,EAAE,CAAC,EAAE,OAAO,EAAE,GAAG,EAAE;YACnE,gBAAgB,EAAE,EAAE,IAAI,EAAE,QAAQ,EAAE,OAAO,EAAE,CAAC,EAAE,OAAO,EAAE,CAAC,EAAE,OAAO,EAAE,IAAI,EAAE;SAC5E;QACD,QAAQ,EAAE,CAAC,MAAM,CAAC;KACnB,CAAC;IAEO,oBAAoB,GAAG,CAAC,gBAAgB,CAAC,CAAC;IAE3C,MAAM,CAAS;IAEvB,YAAY,MAAe;QACzB,IAAI,CAAC,MAAM,GAAG,MAAM,IAAI,OAAO,CAAC,GAAG,CAAC,kBAAkB,IAAI,EAAE,CAAC;IAC/D,CAAC;IAED,KAAK,CAAC,OAAO,CAAC,IAAc,EAAE,QAA8B;QAC1D,IAAI,CAAC,IAAI,CAAC,MAAM;YAAE,OAAO,EAAE,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,oCAAoC,EAAE,CAAC;QAEzF,MAAM,IAAI,GAAG,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC;QACtC,MAAM,OAAO,GAAG,MAAM,CAAC,CAAC,IAAI,CAAC,KAAK,IAAI,QAAQ,CAAC,CAAC,WAAW,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,IAAI,MAAM,CAAC,MAAM,CAAC;QAC9F,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,IAAI,uBAAuB,CAAC;QAEpD,IAAI,CAAC;YACH,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,+CAA+C,OAAO,EAAE,EAAE;gBACrF,MAAM,EAAE,MAAM;gBACd,OAAO,EAAE,EAAE,YAAY,EAAE,IAAI,CAAC,MAAM,EAAE,cAAc,EAAE,kBAAkB,EAAE,MAAM,EAAE,YAAY,EAAE;gBAChG,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC;oBACnB,IAAI;oBACJ,QAAQ,EAAE,KAAK;oBACf,cAAc,EAAE,EAAE,SAAS,EAAE,IAAI,CAAC,SAAS,IAAI,GAAG,EAAE,gBAAgB,EAAE,IAAI,CAAC,gBAAgB,IAAI,IAAI,EAAE;iBACtG,CAAC;aACH,CAAC,CAAC;YAEH,IAAI,CAAC,QAAQ,CAAC,EAAE,EAAE,CAAC;gBACjB,MAAM,GAAG,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,CAAC;gBAClC,OAAO,EAAE,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,qBAAqB,QAAQ,CAAC,MAAM,MAAM,GAAG,EAAE,EAAE,CAAC;YACpF,CAAC;YAED,MAAM,GAAG,GAAG,MAAM,QAAQ,CAAC,WAAW,EAAE,CAAC;YACzC,MAAM,WAAW,GAAG,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,QAAQ,CAAC,QAAQ,CAAC,CAAC;YACxD,MAAM,kBAAkB,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,MAAM,GAAG,GAAG,CAAC,GAAG,EAAE,GAAG,IAAI,CAAC,CAAC;YAEpF,OAAO;gBACL,OAAO,EAAE,IAAI;gBACb,MAAM,EAAE,EAAE,IAAI,EAAE,KAAK,EAAE,IAAI,CAAC,KAAK,IAAI,QAAQ,EAAE,KAAK,EAAE,WAAW,EAAE,WAAW,EAAE,YAAY,EAAE,kBAAkB,EAAE;gBAClH,WAAW,EAAE,YAAY;aAC1B,CAAC;QACJ,CAAC;QAAC,OAAO,GAAQ,EAAE,CAAC;YAClB,OAAO,EAAE,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,eAAe,GAAG,CAAC,OAAO,EAAE,EAAE,CAAC;QACjE,CAAC;IACH,CAAC;CACF"}

package/manifest.json

@@ -0,0 +1,20 @@
+{
+  "$schema": "https://agentos.sh/schemas/extension-manifest-v1.json",
+  "id": "com.framers.media.voice-synthesis",
+  "name": "Voice Synthesis Extension",
+  "version": "1.0.0",
+  "description": "Text-to-speech synthesis via ElevenLabs for generating agent voice audio",
+  "author": { "name": "Frame.dev" },
+  "license": "MIT",
+  "keywords": ["tts", "voice", "speech", "elevenlabs", "audio"],
+  "agentosVersion": "^2.0.0",
+  "categories": ["media", "communication"],
+  "extensions": [
+    { "kind": "tool", "id": "textToSpeech", "displayName": "Text to Speech", "entry": "./src/tools/textToSpeech.ts" }
+  ],
+  "configuration": {
+    "properties": {
+      "elevenlabs.apiKey": { "type": "string", "secret": true }
+    }
+  }
+}

package/package.json

@@ -0,0 +1,38 @@
+{
+  "name": "@framers/agentos-ext-voice-synthesis",
+  "version": "1.0.0",
+  "description": "ElevenLabs text-to-speech synthesis tool for AgentOS",
+  "main": "dist/index.js",
+  "types": "dist/index.d.ts",
+  "type": "module",
+  "sideEffects": false,
+  "exports": {
+    ".": { "import": "./dist/index.js", "types": "./dist/index.d.ts" },
+    "./manifest.json": "./manifest.json",
+    "./package.json": "./package.json"
+  },
+  "scripts": {
+    "build": "tsc",
+    "test": "vitest run",
+    "test:watch": "vitest",
+    "test:coverage": "vitest run --coverage",
+    "lint": "eslint src --ext .ts",
+    "typecheck": "tsc --noEmit",
+    "clean": "rimraf dist"
+  },
+  "keywords": ["agentos", "extension", "tts", "voice", "elevenlabs", "speech", "media"],
+  "author": { "name": "Framers AI", "email": "team@frame.dev", "url": "https://frame.dev" },
+  "contributors": [{ "name": "Johnny Dunn", "email": "johnnyfived@protonmail.com", "url": "https://github.com/jddunn" }],
+  "license": "MIT",
+  "repository": { "type": "git", "url": "https://github.com/framersai/agentos-extensions.git", "directory": "registry/curated/media/voice-synthesis" },
+  "publishConfig": { "access": "public" },
+  "peerDependencies": { "@framers/agentos": "^0.1.0" },
+  "devDependencies": {
+    "@framers/agentos": "^0.1.0",
+    "@types/node": "^20.12.12",
+    "@vitest/coverage-v8": "^1.6.0",
+    "rimraf": "^5.0.7",
+    "typescript": "^5.4.5",
+    "vitest": "^1.6.0"
+  }
+}

package/src/index.ts

@@ -0,0 +1,30 @@
+/**
+ * Voice Synthesis Extension Pack — ElevenLabs TTS for agents.
+ */
+
+import { TextToSpeechTool } from './tools/textToSpeech.js';
+
+export interface VoiceSynthesisExtensionOptions {
+  elevenLabsApiKey?: string;
+  priority?: number;
+}
+
+export function createExtensionPack(context: any) {
+  const options = (context.options || {}) as VoiceSynthesisExtensionOptions;
+  const apiKey = options.elevenLabsApiKey || context.getSecret?.('elevenlabs.apiKey') || process.env.ELEVENLABS_API_KEY;
+  const tool = new TextToSpeechTool(apiKey);
+
+  return {
+    name: '@framers/agentos-ext-voice-synthesis',
+    version: '1.0.0',
+    descriptors: [
+      { id: 'textToSpeech', kind: 'tool' as const, priority: options.priority || 50, payload: tool, requiredSecrets: [{ id: 'elevenlabs.apiKey' }] },
+    ],
+    onActivate: async () => context.logger?.info('Voice Synthesis Extension activated'),
+    onDeactivate: async () => context.logger?.info('Voice Synthesis Extension deactivated'),
+  };
+}
+
+export { TextToSpeechTool };
+export type { TTSInput, TTSOutput } from './tools/textToSpeech.js';
+export default createExtensionPack;

package/src/tools/textToSpeech.ts

@@ -0,0 +1,102 @@
+/**
+ * ElevenLabs TTS Tool — text-to-speech synthesis.
+ */
+
+import type { ITool, ToolExecutionContext, ToolExecutionResult, JSONSchemaObject } from '../../../../../../agentos/src/core/tools/ITool.js';
+
+export interface TTSInput {
+  text: string;
+  voice?: string;
+  model?: string;
+  stability?: number;
+  similarity_boost?: number;
+}
+
+export interface TTSOutput {
+  text: string;
+  voice: string;
+  model: string;
+  audioBase64: string;
+  contentType: string;
+  durationEstimateMs: number;
+}
+
+const VOICES: Record<string, string> = {
+  rachel: '21m00Tcm4TlvDq8ikWAM',
+  domi: 'AZnzlk1XvdvUeBnXmlld',
+  bella: 'EXAVITQu4vr4xnSDxMaL',
+  antoni: 'ErXwobaYiN019PkySvjV',
+  josh: 'TxGEqnHWrfWFTfGW9XjX',
+  arnold: 'VR6AewLTigWG4xSOukaG',
+  adam: 'pNInz6obpgDQGcFmaJgB',
+  sam: 'yoZ06aMxZJJ28mfd3POQ',
+};
+
+export class TextToSpeechTool implements ITool<TTSInput, TTSOutput> {
+  readonly id = 'elevenlabs-tts-v1';
+  readonly name = 'text_to_speech';
+  readonly displayName = 'Text to Speech';
+  readonly description =
+    'Convert text to speech using ElevenLabs. Returns base64-encoded MP3 audio. ' +
+    'Voices: rachel, domi, bella, antoni, josh, arnold, adam, sam.';
+  readonly category = 'media';
+  readonly version = '1.0.0';
+  readonly hasSideEffects = false;
+
+  readonly inputSchema: JSONSchemaObject = {
+    type: 'object',
+    properties: {
+      text: { type: 'string', description: 'Text to convert. Max 5000 chars.' },
+      voice: { type: 'string', default: 'rachel', description: 'Voice name or ID.' },
+      model: { type: 'string', default: 'eleven_monolingual_v1' },
+      stability: { type: 'number', minimum: 0, maximum: 1, default: 0.5 },
+      similarity_boost: { type: 'number', minimum: 0, maximum: 1, default: 0.75 },
+    },
+    required: ['text'],
+  };
+
+  readonly requiredCapabilities = ['capability:tts'];
+
+  private apiKey: string;
+
+  constructor(apiKey?: string) {
+    this.apiKey = apiKey || process.env.ELEVENLABS_API_KEY || '';
+  }
+
+  async execute(args: TTSInput, _context: ToolExecutionContext): Promise<ToolExecutionResult<TTSOutput>> {
+    if (!this.apiKey) return { success: false, error: 'ELEVENLABS_API_KEY not configured.' };
+
+    const text = args.text.slice(0, 5000);
+    const voiceId = VOICES[(args.voice || 'rachel').toLowerCase()] || args.voice || VOICES.rachel;
+    const model = args.model || 'eleven_monolingual_v1';
+
+    try {
+      const response = await fetch(`https://api.elevenlabs.io/v1/text-to-speech/${voiceId}`, {
+        method: 'POST',
+        headers: { 'xi-api-key': this.apiKey, 'Content-Type': 'application/json', Accept: 'audio/mpeg' },
+        body: JSON.stringify({
+          text,
+          model_id: model,
+          voice_settings: { stability: args.stability ?? 0.5, similarity_boost: args.similarity_boost ?? 0.75 },
+        }),
+      });
+
+      if (!response.ok) {
+        const err = await response.text();
+        return { success: false, error: `ElevenLabs error (${response.status}): ${err}` };
+      }
+
+      const buf = await response.arrayBuffer();
+      const audioBase64 = Buffer.from(buf).toString('base64');
+      const durationEstimateMs = Math.round((text.split(/\s+/).length / 150) * 60 * 1000);
+
+      return {
+        success: true,
+        output: { text, voice: args.voice || 'rachel', model, audioBase64, contentType: 'audio/mpeg', durationEstimateMs },
+        contentType: 'audio/mpeg',
+      };
+    } catch (err: any) {
+      return { success: false, error: `TTS failed: ${err.message}` };
+    }
+  }
+}

package/test/textToSpeech.spec.ts

@@ -0,0 +1,97 @@
+import { describe, it, expect, vi, beforeEach } from 'vitest';
+
+const mockFetch = vi.fn();
+vi.stubGlobal('fetch', mockFetch);
+
+const { TextToSpeechTool } = await import('../src/tools/textToSpeech.js');
+const { createExtensionPack } = await import('../src/index.js');
+
+describe('TextToSpeechTool', () => {
+  let tool: InstanceType<typeof TextToSpeechTool>;
+
+  beforeEach(() => {
+    vi.clearAllMocks();
+    tool = new TextToSpeechTool('test-api-key');
+  });
+
+  describe('metadata', () => {
+    it('has correct id and name', () => {
+      expect(tool.id).toBe('elevenlabs-tts-v1');
+      expect(tool.name).toBe('text_to_speech');
+    });
+
+    it('has valid input schema', () => {
+      expect(tool.inputSchema.type).toBe('object');
+      expect(tool.inputSchema.required).toContain('text');
+    });
+
+    it('has no side effects', () => {
+      expect(tool.hasSideEffects).toBe(false);
+    });
+  });
+
+  describe('execute', () => {
+    const ctx = {} as any;
+
+    it('returns error when API key is missing', async () => {
+      const noKeyTool = new TextToSpeechTool('');
+      const result = await noKeyTool.execute({ text: 'Hello' }, ctx);
+      expect(result.success).toBe(false);
+      expect(result.error).toContain('ELEVENLABS_API_KEY');
+    });
+
+    it('synthesizes speech successfully', async () => {
+      const mockAudio = new ArrayBuffer(100);
+      mockFetch.mockResolvedValueOnce({
+        ok: true,
+        arrayBuffer: async () => mockAudio,
+      });
+
+      const result = await tool.execute({ text: 'Hello world' }, ctx);
+      expect(result.success).toBe(true);
+      expect(result.output!.voice).toBe('rachel');
+      expect(result.output!.contentType).toBe('audio/mpeg');
+      expect(result.output!.audioBase64).toBeTruthy();
+      expect(result.output!.durationEstimateMs).toBeGreaterThan(0);
+    });
+
+    it('uses correct voice ID for named voice', async () => {
+      mockFetch.mockResolvedValueOnce({ ok: true, arrayBuffer: async () => new ArrayBuffer(10) });
+      await tool.execute({ text: 'Test', voice: 'josh' }, ctx);
+      expect(mockFetch).toHaveBeenCalledWith(
+        expect.stringContaining('TxGEqnHWrfWFTfGW9XjX'),
+        expect.any(Object)
+      );
+    });
+
+    it('truncates text to 5000 chars', async () => {
+      mockFetch.mockResolvedValueOnce({ ok: true, arrayBuffer: async () => new ArrayBuffer(10) });
+      const longText = 'a'.repeat(6000);
+      const result = await tool.execute({ text: longText }, ctx);
+      expect(result.success).toBe(true);
+      expect(result.output!.text.length).toBe(5000);
+    });
+
+    it('handles API errors', async () => {
+      mockFetch.mockResolvedValueOnce({ ok: false, status: 401, text: async () => 'Unauthorized' });
+      const result = await tool.execute({ text: 'Test' }, ctx);
+      expect(result.success).toBe(false);
+      expect(result.error).toContain('401');
+    });
+
+    it('handles network errors', async () => {
+      mockFetch.mockRejectedValueOnce(new Error('Timeout'));
+      const result = await tool.execute({ text: 'Test' }, ctx);
+      expect(result.success).toBe(false);
+      expect(result.error).toContain('Timeout');
+    });
+  });
+});
+
+describe('createExtensionPack', () => {
+  it('creates pack with correct metadata', () => {
+    const pack = createExtensionPack({ options: { elevenLabsApiKey: 'test' }, logger: { info: vi.fn() } });
+    expect(pack.name).toBe('@framers/agentos-ext-voice-synthesis');
+    expect(pack.descriptors).toHaveLength(1);
+  });
+});

package/tsconfig.json

@@ -0,0 +1,22 @@
+{
+  "compilerOptions": {
+    "target": "ES2022",
+    "module": "ESNext",
+    "lib": ["ES2022"],
+    "outDir": "./dist",
+    "rootDir": "./src",
+    "strict": true,
+    "esModuleInterop": true,
+    "skipLibCheck": true,
+    "noEmitOnError": false,
+    "forceConsistentCasingInFileNames": true,
+    "declaration": true,
+    "declarationMap": true,
+    "sourceMap": true,
+    "resolveJsonModule": true,
+    "moduleResolution": "Bundler",
+    "types": ["node"]
+  },
+  "include": ["src/**/*"],
+  "exclude": ["node_modules", "dist", "test"]
+}

package/vitest.config.ts

@@ -0,0 +1,10 @@
+import { defineConfig } from 'vitest/config';
+
+export default defineConfig({
+  test: {
+    globals: true,
+    environment: 'node',
+    include: ['test/**/*.spec.ts'],
+    testTimeout: 10000,
+  },
+});