Merge pull request #155 from dcSpark/feature/add-macos-say-text-to-au…

…dio-tool

feat: add macos-say-text-to-audio tool

tools/macos-say-text-to-audio/index.test.ts

@@ -0,0 +1,42 @@
+import { expect } from '@jest/globals';
+import { getToolTestClient } from '../../src/test/utils';
+import * as path from 'path';
+
+describe('macOS Text-to-Speech', () => {
+  const toolPath = path.join(__dirname, 'tool.py');
+  const client = getToolTestClient();
+
+  it('speaks text with default voice and rate', async () => {
+    const response = await client.executeToolFromFile(
+      toolPath,
+      { text: "Hello from text to speech test" },
+      {},
+      []
+    );
+
+    expect(response).toHaveProperty('success', true);
+    expect(response.message).toMatch(/Spoke text with voice=Alex/);
+  }, 15000);
+
+  it('handles empty text gracefully', async () => {
+    await expect(
+      client.executeToolFromFile(toolPath, { text: '' }, {}, [])
+    ).rejects.toThrow(/No text provided/);
+  });
+
+  it('respects custom voice and rate', async () => {
+    const response = await client.executeToolFromFile(
+      toolPath,
+      {
+        text: "Testing custom voice and rate",
+        voice: "Daniel",
+        rate: 200
+      },
+      {},
+      []
+    );
+    expect(response.success).toBe(true);
+    expect(response.message).toMatch(/voice=Daniel/);
+    expect(response.message).toMatch(/rate=200/);
+  }, 15000);
+}); 

tools/macos-say-text-to-audio/metadata.json

@@ -0,0 +1,51 @@
+{
+  "id": "macos-say-text-to-audio",
+  "version": "1.0.0",
+  "name": "macOS Text-to-Speech",
+  "description": "Speaks text aloud using macOS 'say' command. Available voices can be found by running 'say -v ?' in terminal, common voices include: Alex (default), Daniel, Karen, Samantha, Victoria",
+  "author": "Shinkai",
+  "keywords": [
+    "macos",
+    "say",
+    "text-to-speech",
+    "audio"
+  ],
+  "configurations": {
+    "type": "object",
+    "properties": {},
+    "required": []
+  },
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "text": {
+        "type": "string",
+        "description": "The text to convert to speech"
+      },
+      "voice": {
+        "type": "string",
+        "description": "The voice to use (e.g., Alex, Daniel, Karen)",
+        "default": "Alex"
+      }
+    },
+    "required": ["text"]
+  },
+  "result": {
+    "type": "object",
+    "properties": {
+      "success": {
+        "type": "boolean",
+        "description": "Whether the text was successfully converted to speech"
+      },
+      "message": {
+        "type": "string",
+        "description": "Success or error message"
+      }
+    },
+    "required": ["success", "message"]
+  },
+  "sqlTables": [],
+  "sqlQueries": [],
+  "tools": [
+  ]
+}    

tools/macos-say-text-to-audio/store.json

@@ -0,0 +1,3 @@
+{
+  "categoryId": "5f10d0b4-6acd-477a-96e1-be35634465b2"
+}

tools/macos-say-text-to-audio/tool.py

@@ -0,0 +1,65 @@
+# /// script
+# dependencies = [
+#   "requests"
+# ]
+# ///
+
+import subprocess
+import sys
+from typing import Dict, Any, Optional, List
+
+class CONFIG:
+    """
+    No configurations needed for this tool by default.
+    But you could expand if you need environment checks, etc.
+    """
+    pass
+
+class INPUTS:
+    text: str  # The text to speak
+    voice: str = "Alex"  # A valid macOS voice, e.g. "Alex", "Daniel", "Victoria", etc.
+    rate: int = 175      # Words per minute. Mac `say` defaults around 175
+    background: bool = False
+
+class OUTPUT:
+    success: bool
+    message: str
+
+async def run(c: CONFIG, p: INPUTS) -> OUTPUT:
+    """
+    Speaks out the given text on macOS using the 'say' command.
+    """
+    # Validate platform
+    if sys.platform != 'darwin':
+        raise RuntimeError("This tool only works on macOS systems")
+
+    # Validate inputs
+    text = p.text.strip()
+    if not text:
+        raise ValueError("No text provided to speak.")
+    if p.rate < 1 or p.rate > 500:
+        raise ValueError("Rate must be between 1 and 500 words per minute.")
+
+    # Build shell command
+    # Example: say -v "Alex" -r 175 "Hello world"
+    # If background is True, add an ampersand at the end (shell background).
+    voice_arg = f'-v "{p.voice}"'
+    rate_arg = f'-r {p.rate}'
+    quoted_text = text.replace('"', '\\"')  # Escape double quotes
+    cmd = f'say {voice_arg} {rate_arg} "{quoted_text}"'
+    if p.background:
+        cmd += " &"
+
+    # Run the command
+    try:
+        subprocess.run(cmd, shell=True, check=True)
+        success = True
+        message = f"Spoke text with voice={p.voice}, rate={p.rate}, background={p.background}"
+    except subprocess.CalledProcessError as e:
+        success = False
+        message = f"Failed to speak text: {str(e)}"
+
+    output = OUTPUT()
+    output.success = success
+    output.message = message
+    return output