Show HN: Give your AI agent a personality (and a voice) without external APIs
The article presents an open‑source tool that adds voice output and selectable personalities to AI agents without relying on external APIs. It provides step‑by‑step instructions for cloning the repository, running a setup script, and configuring the voice server across various environments. Users can choose from several predefined personality profiles to alter the agent’s spoken behavior.
- ▪The repository can be cloned and initialized with a Python setup script that automatically configures persona, permissions, and a virtual environment.
- ▪Manual configuration is supported via JSON files for platforms such as Google Antigravity, Claude CLI, Cursor IDE, and Windsurf Editor, specifying the voice command and arguments.
- ▪The tool offers multiple personality profiles—including Sarcastic Senior, Eager Intern, Existential Emo, and others—that modify how the agent speaks.
- ▪All voice functionality runs locally, eliminating the need for external API calls.
Opening excerpt (first ~120 words) tap to expand
Setup Guide Follow these steps to initialize your voice server. 1 Clone Repository Fetch the code and navigate into the project directory. git clone https://github.com/fellowgeek/mcp-speak.git && cd mcp-speak 2 Run Setup Wizard Configures your persona, permissions, and virtual environment automatically. python3 setup.py Manual Integration (Optional) Running python3 setup.py configures your client automatically. Below are manual JSON reference snippets. Google Antigravity Config (~/.gemini/antigravity/mcp_config.json) { "mcpServers": { "voice": { "command": "/ABSOLUTE/PATH/TO/run.sh" } } } Claude CLI Config claude mcp add --scope user voice /ABSOLUTE/PATH/TO/run.sh Claude Desktop Config (~/Library/Application Support/Claude/claude_desktop_config.json) { "mcpServers": { "voice": {…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Hacker News - Newest: ""AI" "LLM"".