Mohamed Boudra ee315bb9a0 ``` ...

refactor: improve session management, audio handling, and MCP configuration

- Change default session name from 'voice-dev' to '__voice-dev'
- Change default server port from 3000 to 6767
- Add conversation history persistence to .conversations directory
- Defer stream abort until after transcription completes
- Return empty content arrays from MCP tool handlers (use structuredContent only)
- Add pause/resume support to audio player for handling false positive speech detection
- Add .conversations to .gitignore
```

2025-10-20 18:14:08 +02:00

.claude

refactor: enhance TTS playback management with abort signal support

2025-10-20 16:19:34 +02:00

packages/voice-assistant

```

2025-10-20 18:14:08 +02:00

.gitignore

Add explicit lightningcss dependency for Vercel build

2025-10-13 14:18:15 +02:00

.tool-versions

Remove unnecessary platform-specific Tailwind CSS Oxide dependencies

2025-10-13 15:00:10 +02:00

package-lock.json

```

2025-10-20 12:47:09 +02:00

package.json

Based on the git changes, this is a major migration from OpenAI's Realtime API to LiveKit for voice infrastructure. Here's the commit message:

2025-10-14 11:54:17 +02:00

tsconfig.base.json

Initial monorepo setup with mcp-server and web packages

2025-10-13 14:03:54 +02:00

Description

Orchestrate multiple coding agents from desktop and mobile

adeagentsandroidclaude-codecodexcopilotdeveloper-toolshermesioslinuxmobileopencodeorchestrationpiwindows

AGPL-3.0 186 MiB

Languages

TypeScript 98.1%

JavaScript 1.3%

Shell 0.1%

Swift 0.1%