agentvibes 4.2.0 → 4.4.0

package/mcp-server/test_windows_script_parity.py

@@ -0,0 +1,336 @@
+#!/usr/bin/env python3
+"""
+File: mcp-server/test_windows_script_parity.py
+
+AgentVibes - Finally, your AI Agents can Talk Back! Text-to-Speech WITH personality for AI Assistants!
+Website: https://agentvibes.org
+Repository: https://github.com/paulpreibisch/AgentVibes
+
+Co-created by Paul Preibisch with Claude AI
+Copyright (c) 2025 Paul Preibisch
+
+Licensed under the Apache License, Version 2.0
+
+---
+
+@fileoverview Windows script parity tests - proves missing scripts break MCP tools
+@context Test-first approach: these tests SHOULD FAIL until Windows scripts are ported
+@architecture pytest-style tests validating script existence and MCP integration
+"""
+
+import os
+import sys
+import platform
+from pathlib import Path
+
+# Add the mcp-server directory to path
+sys.path.insert(0, str(Path(__file__).parent))
+
+IS_WINDOWS = platform.system() == "Windows" and not os.environ.get("WSL_DISTRO_NAME")
+
+# Resolve project paths
+PROJECT_ROOT = Path(__file__).parent.parent
+HOOKS_WINDOWS = PROJECT_ROOT / ".claude" / "hooks-windows"
+HOOKS_LINUX = PROJECT_ROOT / ".claude" / "hooks"
+
+
+# =============================================================
+# Suite 1: Every .sh script called by MCP must have a .ps1
+# =============================================================
+
+# Scripts the MCP server calls via _run_script() with hardcoded .sh names
+# The auto-conversion (server.py:902-903) converts .sh -> .ps1
+MCP_HARDCODED_SCRIPTS = [
+    "provider-manager",
+    "learn-manager",
+    "speed-manager",
+    "download-extra-voices",
+    "verbosity-manager",
+    "clean-audio-cache",
+]
+
+# Scripts set via class constants (already .ps1 on Windows)
+MCP_CLASS_CONSTANT_SCRIPTS = [
+    "voice-manager-windows",   # VOICE_MANAGER_SCRIPT override
+    "personality-manager",     # PERSONALITY_MANAGER_SCRIPT
+    "language-manager",        # LANGUAGE_MANAGER_SCRIPT
+    "background-music-manager",  # BACKGROUND_MUSIC_MANAGER_SCRIPT
+    "effects-manager",         # EFFECTS_MANAGER_SCRIPT
+]
+
+
+def test_hardcoded_scripts_have_ps1():
+    """Every .sh script called via _run_script must have a .ps1 in hooks-windows/"""
+    print("\nTesting hardcoded .sh -> .ps1 script parity...")
+    missing = []
+
+    for script_name in MCP_HARDCODED_SCRIPTS:
+        ps1_path = HOOKS_WINDOWS / f"{script_name}.ps1"
+        sh_path = HOOKS_LINUX / f"{script_name}.sh"
+
+        sh_exists = sh_path.exists()
+        ps1_exists = ps1_path.exists()
+
+        if sh_exists and not ps1_exists:
+            missing.append(script_name)
+            print(f"  FAIL: {script_name}.sh exists but {script_name}.ps1 is MISSING")
+        elif ps1_exists:
+            print(f"  PASS: {script_name}.ps1 exists")
+        else:
+            print(f"  SKIP: {script_name}.sh doesn't exist either")
+
+    if missing:
+        print(f"\n  {len(missing)} missing .ps1 scripts: {', '.join(missing)}")
+    assert len(missing) == 0, (
+        f"Missing Windows scripts for MCP: {', '.join(missing)}. "
+        f"These MCP tools silently fail on Windows."
+    )
+    print("  All hardcoded scripts have .ps1 equivalents")
+    return True
+
+
+def test_class_constant_scripts_exist():
+    """Every script set via class constants must exist in hooks-windows/"""
+    print("\nTesting class constant script existence...")
+    missing = []
+
+    for script_name in MCP_CLASS_CONSTANT_SCRIPTS:
+        ps1_path = HOOKS_WINDOWS / f"{script_name}.ps1"
+        if not ps1_path.exists():
+            missing.append(script_name)
+            print(f"  FAIL: {script_name}.ps1 is MISSING")
+        else:
+            print(f"  PASS: {script_name}.ps1 exists")
+
+    if missing:
+        print(f"\n  {len(missing)} missing .ps1 scripts: {', '.join(missing)}")
+    assert len(missing) == 0, (
+        f"Missing Windows scripts (class constants): {', '.join(missing)}. "
+        f"These MCP tools fail on Windows."
+    )
+    print("  All class constant scripts exist")
+    return True
+
+
+# =============================================================
+# Suite 2: MCP _run_script returns proper errors for missing scripts
+# =============================================================
+
+def test_run_script_error_for_missing():
+    """Verify _run_script returns 'Script not found' for missing scripts (Windows only)"""
+    if not IS_WINDOWS:
+        print("\nSkipping _run_script test (not Windows)")
+        return True
+
+    print("\nTesting _run_script error handling for missing scripts...")
+    try:
+        from server import AgentVibesServer
+        import asyncio
+
+        server = AgentVibesServer()
+
+        async def check_missing():
+            # Try calling a script we know doesn't exist
+            result = await server._run_script("definitely-does-not-exist.ps1", [])
+            assert "Script not found" in result, (
+                f"Expected 'Script not found' for missing script, got: {result}"
+            )
+            print("  PASS: _run_script returns 'Script not found' for missing scripts")
+
+        asyncio.run(check_missing())
+        return True
+
+    except ImportError:
+        print("  SKIP: MCP library not installed")
+        return True
+    except Exception as e:
+        print(f"  FAIL: {e}")
+        return False
+
+
+# =============================================================
+# Suite 3: MCP tools that depend on missing scripts
+# =============================================================
+
+MCP_TOOL_SCRIPT_MAP = {
+    "set_speed": "speed-manager",
+    "get_speed": "speed-manager",
+    "set_learn_mode": "learn-manager",
+    "set_verbosity": "verbosity-manager",
+    "get_verbosity": "verbosity-manager",
+    "mute": "verbosity-manager",
+    "unmute": "verbosity-manager",
+    "is_muted": "verbosity-manager",
+    "set_personality": "personality-manager",
+    "list_personalities": "personality-manager",
+    "set_language": "language-manager",
+    "clean_audio_cache": "clean-audio-cache",
+}
+
+
+def test_mcp_tools_have_backing_scripts():
+    """Every MCP tool must have its backing script on Windows"""
+    print("\nTesting MCP tool -> script mapping on Windows...")
+    broken_tools = []
+
+    for tool_name, script_name in MCP_TOOL_SCRIPT_MAP.items():
+        ps1_path = HOOKS_WINDOWS / f"{script_name}.ps1"
+        if not ps1_path.exists():
+            broken_tools.append((tool_name, script_name))
+            print(f"  FAIL: MCP tool '{tool_name}' -> {script_name}.ps1 MISSING")
+        else:
+            print(f"  PASS: MCP tool '{tool_name}' -> {script_name}.ps1 exists")
+
+    if broken_tools:
+        tool_list = ", ".join(f"{t[0]} ({t[1]}.ps1)" for t in broken_tools)
+        print(f"\n  {len(broken_tools)} broken MCP tools: {tool_list}")
+    assert len(broken_tools) == 0, (
+        f"Broken MCP tools on Windows: "
+        + ", ".join(f"{t[0]}" for t in broken_tools)
+    )
+    print("  All MCP tools have backing scripts")
+    return True
+
+
+# =============================================================
+# Suite 4: Provider naming validation
+# =============================================================
+
+def test_provider_names_consistent():
+    """Verify MCP server accepts unified provider names"""
+    if not IS_WINDOWS:
+        print("\nSkipping provider naming test (not Windows)")
+        return True
+
+    print("\nTesting provider naming consistency...")
+    try:
+        from server import AgentVibesServer
+
+        server = AgentVibesServer()
+
+        # The MCP server hardcodes different valid providers per platform
+        # On Windows: ["windows-piper", "windows-sapi", "soprano"]
+        # The provider-manager.ps1 also accepts: ["piper", "sapi"]
+        # This creates a mismatch where scripts accept names MCP rejects
+
+        # Read the valid_providers from set_provider source
+        import inspect
+        source = inspect.getsource(server.set_provider)
+
+        if "windows-piper" in source and "piper" not in source.replace("windows-piper", ""):
+            print("  INFO: MCP only accepts 'windows-piper', not 'piper' on Windows")
+            print("  INFO: But provider-manager.ps1 accepts both")
+            print("  WARN: This creates a one-way trap - consider unifying to 'piper'")
+
+        # Check that play-tts scripts match provider names
+        for provider in ["piper", "sapi", "soprano"]:
+            windows_name = f"play-tts-windows-{provider}.ps1"
+            unified_name = f"play-tts-{provider}.ps1"
+            has_windows = (HOOKS_WINDOWS / windows_name).exists()
+            has_unified = (HOOKS_WINDOWS / unified_name).exists()
+
+            if has_windows and not has_unified:
+                print(f"  WARN: {windows_name} exists but {unified_name} does not")
+                print(f"         Auto-conversion from play-tts-{provider}.sh would look for {unified_name}")
+
+        print("  Provider naming test complete (warnings above need fixing)")
+        return True
+
+    except ImportError:
+        print("  SKIP: MCP library not installed")
+        return True
+    except Exception as e:
+        print(f"  FAIL: {e}")
+        return False
+
+
+# =============================================================
+# Suite 5: Voice manager naming mismatch
+# =============================================================
+
+def test_voice_manager_naming():
+    """voice-manager.sh auto-converts to voice-manager.ps1 but file is voice-manager-windows.ps1"""
+    print("\nTesting voice-manager naming consistency...")
+
+    # The auto-conversion would produce voice-manager.ps1
+    auto_converted = HOOKS_WINDOWS / "voice-manager.ps1"
+    actual_file = HOOKS_WINDOWS / "voice-manager-windows.ps1"
+    linux_file = HOOKS_LINUX / "voice-manager.sh"
+
+    print(f"  Linux script: voice-manager.sh exists = {linux_file.exists()}")
+    print(f"  Auto-converted name: voice-manager.ps1 exists = {auto_converted.exists()}")
+    print(f"  Actual Windows file: voice-manager-windows.ps1 exists = {actual_file.exists()}")
+
+    if actual_file.exists() and not auto_converted.exists():
+        print("  WARN: Naming mismatch detected!")
+        print("        MCP class constant overrides this, but it's fragile")
+        # This is informational, not a hard failure since the class constant handles it
+        # But it should be fixed for consistency
+
+    if linux_file.exists():
+        assert auto_converted.exists() or actual_file.exists(), (
+            "voice-manager has no Windows equivalent at all"
+        )
+
+    print("  Voice manager naming test complete")
+    return True
+
+
+# =============================================================
+# Main
+# =============================================================
+
+def main():
+    """Run all parity tests"""
+    print("=" * 60)
+    print("AgentVibes Windows Script Parity Tests")
+    print(f"Platform: {platform.system()} (IS_WINDOWS={IS_WINDOWS})")
+    print(f"hooks-windows/: {HOOKS_WINDOWS}")
+    print("=" * 60)
+
+    tests = [
+        ("Hardcoded .sh scripts have .ps1 equivalents", test_hardcoded_scripts_have_ps1),
+        ("Class constant scripts exist", test_class_constant_scripts_exist),
+        ("_run_script error handling", test_run_script_error_for_missing),
+        ("MCP tools have backing scripts", test_mcp_tools_have_backing_scripts),
+        ("Provider naming consistency", test_provider_names_consistent),
+        ("Voice manager naming", test_voice_manager_naming),
+    ]
+
+    results = []
+    for name, test_func in tests:
+        try:
+            result = test_func()
+            results.append((name, result, None))
+        except AssertionError as e:
+            print(f"\n  ASSERTION FAILED: {e}")
+            results.append((name, False, str(e)))
+        except Exception as e:
+            print(f"\n  ERROR: {e}")
+            results.append((name, False, str(e)))
+
+    print("\n" + "=" * 60)
+    print("PARITY TEST RESULTS")
+    print("=" * 60)
+
+    passed = sum(1 for _, result, _ in results if result)
+    failed = sum(1 for _, result, _ in results if not result)
+
+    for name, result, error in results:
+        status = "PASS" if result else "FAIL"
+        print(f"  [{status}] {name}")
+        if error:
+            print(f"         {error[:100]}")
+
+    print("=" * 60)
+    print(f"Passed: {passed}/{len(results)}  Failed: {failed}/{len(results)}")
+
+    if failed > 0:
+        print(f"\n{failed} test(s) failed — these represent missing Windows scripts")
+        print("See GitHub issue #157 for the full platform parity plan")
+
+    return 1 if failed > 0 else 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

package/package.json

@@ -1,110 +1,110 @@
-{
-  "$schema": "https://json.schemastore.org/package.json",
-  "name": "agentvibes",
-  "version": "4.2.0",
-  "description": "Now your AI Agents can finally talk back! Professional TTS voice for Claude Code, Claude Desktop (via MCP), and Clawdbot with multi-provider support.",
-  "homepage": "https://agentvibes.org",
-  "keywords": [
-    "tts",
-    "text-to-speech",
-    "piper-tts",
-    "claude-code",
-    "claude-desktop",
-    "clawdbot",
-    "mcp",
-    "model-context-protocol",
-    "voice",
-    "ai",
-    "narration",
-    "agent-vibes"
-  ],
-  "repository": {
-    "type": "git",
-    "url": "git+https://github.com/paulpreibisch/AgentVibes.git"
-  },
-  "license": "Apache-2.0",
-  "author": "Paul Preibisch <paul@paulpreibisch.com>",
-  "type": "module",
-  "main": "src/installer.js",
-  "bin": {
-    "agentvibes": "bin/agentvibes.js",
-    "agent-vibes": "bin/agent-vibes",
-    "agentvibes-mcp-server": "bin/mcp-server.js",
-    "agentvibes-voice-browser": "bin/agentvibes-voice-browser.js",
-    "test-bmad-pr": "bin/test-bmad-pr"
-  },
-  "files": [
-    "bin/",
-    "src/",
-    "templates/*.sh",
-    "templates/*.md",
-    "templates/audio/*.mp3",
-    "mcp-server/*.py",
-    "mcp-server/*.js",
-    "mcp-server/*.md",
-    "mcp-server/*.txt",
-    "mcp-server/*.toml",
-    "mcp-server/*.json",
-    "mcp-server/docs/",
-    "mcp-server/examples/",
-    ".claude/commands/agent-vibes/",
-    ".claude/commands/agent-vibes-bmad-voices.md",
-    ".claude/commands/agent-vibes-rdp.md",
-    ".claude/hooks/",
-    ".claude/hooks-windows/",
-    ".claude/personalities/",
-    ".claude/output-styles/",
-    ".claude/audio/tracks/",
-    ".claude/config/",
-    ".claude/docs/",
-    ".claude/language-voices.yaml",
-    ".claude/settings.json",
-    ".claude/activation-instructions",
-    ".claude/piper-voices-dir.txt",
-    ".claude/verbosity.txt",
-    ".claude/github-star-reminder.txt",
-    ".claude/audio/voice-samples/",
-    "voice-assignments.json",
-    ".agentvibes/",
-    ".clawdbot/",
-    ".mcp.json",
-    "setup-windows.ps1",
-    "WINDOWS-SETUP.md",
-    "README.md",
-    "RELEASE_NOTES.md",
-    "LICENSE",
-    "CLAUDE.md"
-  ],
-  "scripts": {
-    "install-local": "node src/installer.js install",
-    "postinstall": "node mcp-server/install-deps.js",
-    "install-mcp-deps": "node mcp-server/install-deps.js",
-    "voice-browser": "node bin/agentvibes-voice-browser.js",
-    "test": "npm run test:syntax && AGENTVIBES_TEST_MODE=true bats test/unit/*.bats && npm run test:coverage",
-    "test:syntax": "node -c src/installer.js && node -c mcp-server/install-deps.js",
-    "test:bats": "AGENTVIBES_TEST_MODE=true bats test/unit/*.bats",
-    "test:node": "node --test test/unit/*.test.js",
-    "test:coverage": "c8 --reporter=lcov --reporter=text node --test test/unit/*.test.js",
-    "test:verbose": "AGENTVIBES_TEST_MODE=true bats -t test/unit/*.bats"
-  },
-  "dependencies": {
-    "@inquirer/search": "^3.1.3",
-    "blessed": "^0.1.81",
-    "boxen": "^7.0.0",
-    "chalk": "^5.0.0",
-    "commander": "^10.0.0",
-    "figlet": "^1.6.0",
-    "inquirer": "^12.0.0",
-    "js-yaml": "^4.1.1",
-    "ora": "^6.0.0"
-  },
-  "engines": {
-    "node": ">=16.0.0"
-  },
-  "publishConfig": {
-    "access": "public"
-  },
-  "devDependencies": {
-    "c8": "^10.1.3"
-  }
-}
+{
+  "$schema": "https://json.schemastore.org/package.json",
+  "name": "agentvibes",
+  "version": "4.4.0",
+  "description": "Now your AI Agents can finally talk back! Professional TTS voice for Claude Code, Claude Desktop (via MCP), and Clawdbot with multi-provider support.",
+  "homepage": "https://agentvibes.org",
+  "keywords": [
+    "tts",
+    "text-to-speech",
+    "piper-tts",
+    "claude-code",
+    "claude-desktop",
+    "clawdbot",
+    "mcp",
+    "model-context-protocol",
+    "voice",
+    "ai",
+    "narration",
+    "agent-vibes"
+  ],
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/paulpreibisch/AgentVibes.git"
+  },
+  "license": "Apache-2.0",
+  "author": "Paul Preibisch <paul@paulpreibisch.com>",
+  "type": "module",
+  "main": "src/installer.js",
+  "bin": {
+    "agentvibes": "bin/agentvibes.js",
+    "agent-vibes": "bin/agent-vibes",
+    "agentvibes-mcp-server": "bin/mcp-server.js",
+    "agentvibes-voice-browser": "bin/agentvibes-voice-browser.js",
+    "test-bmad-pr": "bin/test-bmad-pr"
+  },
+  "files": [
+    "bin/",
+    "src/",
+    "templates/*.sh",
+    "templates/*.md",
+    "templates/audio/*.mp3",
+    "mcp-server/*.py",
+    "mcp-server/*.js",
+    "mcp-server/*.md",
+    "mcp-server/*.txt",
+    "mcp-server/*.toml",
+    "mcp-server/*.json",
+    "mcp-server/docs/",
+    "mcp-server/examples/",
+    ".claude/commands/agent-vibes/",
+    ".claude/commands/agent-vibes-bmad-voices.md",
+    ".claude/commands/agent-vibes-rdp.md",
+    ".claude/hooks/",
+    ".claude/hooks-windows/",
+    ".claude/personalities/",
+    ".claude/output-styles/",
+    ".claude/audio/tracks/",
+    ".claude/config/",
+    ".claude/docs/",
+    ".claude/language-voices.yaml",
+    ".claude/settings.json",
+    ".claude/activation-instructions",
+    ".claude/piper-voices-dir.txt",
+    ".claude/verbosity.txt",
+    ".claude/github-star-reminder.txt",
+    ".claude/audio/voice-samples/",
+    "voice-assignments.json",
+    ".agentvibes/",
+    ".clawdbot/",
+    ".mcp.json",
+    "setup-windows.ps1",
+    "WINDOWS-SETUP.md",
+    "README.md",
+    "RELEASE_NOTES.md",
+    "LICENSE",
+    "CLAUDE.md"
+  ],
+  "scripts": {
+    "install-local": "node src/installer.js install",
+    "postinstall": "node mcp-server/install-deps.js",
+    "install-mcp-deps": "node mcp-server/install-deps.js",
+    "voice-browser": "node bin/agentvibes-voice-browser.js",
+    "test": "npm run test:syntax && AGENTVIBES_TEST_MODE=true bats test/unit/*.bats && npm run test:coverage",
+    "test:syntax": "node -c src/installer.js && node -c mcp-server/install-deps.js",
+    "test:bats": "AGENTVIBES_TEST_MODE=true bats test/unit/*.bats",
+    "test:node": "node --test test/unit/*.test.js",
+    "test:coverage": "c8 --reporter=lcov --reporter=text node --test test/unit/*.test.js",
+    "test:verbose": "AGENTVIBES_TEST_MODE=true bats -t test/unit/*.bats"
+  },
+  "dependencies": {
+    "@inquirer/search": "^3.1.3",
+    "blessed": "^0.1.81",
+    "boxen": "^7.0.0",
+    "chalk": "^5.0.0",
+    "commander": "^10.0.0",
+    "figlet": "^1.6.0",
+    "inquirer": "^12.0.0",
+    "js-yaml": "^4.1.1",
+    "ora": "^6.0.0"
+  },
+  "engines": {
+    "node": ">=16.0.0"
+  },
+  "publishConfig": {
+    "access": "public"
+  },
+  "devDependencies": {
+    "c8": "^10.1.3"
+  }
+}