Vox MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| XAI_API_KEY | No | API key for xAI | |
| DEFAULT_MODEL | No | Default model, default 'auto' | |
| CUSTOM_API_URL | No | Custom API URL for Ollama, vLLM, etc. | |
| GEMINI_API_KEY | No | API key for Google Gemini | |
| OPENAI_API_KEY | No | API key for OpenAI | |
| DEEPSEEK_API_KEY | No | API key for DeepSeek | |
| MOONSHOT_API_KEY | No | API key for Moonshot (Kimi) | |
| ANTHROPIC_API_KEY | No | API key for Anthropic | |
| OPENROUTER_API_KEY | No | API key for OpenRouter | |
| MAX_CONVERSATION_TURNS | No | Thread length limit, default 100 | |
| VOX_FORCE_ENV_OVERRIDE | No | Force .env override of client env vars | |
| CONVERSATION_TIMEOUT_HOURS | No | Thread TTL in hours, default 24 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| chatA | Multi-model AI gateway. Routes prompts to external AI models (Gemini, OpenAI, Anthropic, DeepSeek, Moonshot, xAI, OpenRouter, custom endpoints) with conversation memory. Supports file context embedding, images, and multi-turn threads via continuation_id. |
| listmodelsA | Shows which AI model providers are configured, available model names, their aliases and capabilities. |
| dump_threadsA | Export conversation threads as JSON or Markdown. Threads persist to disk and can be cold-reloaded after memory expiry. Use thread_ids to filter specific threads, format to choose output. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: chat for conversation, listmodels for model configuration, and dump_threads for exporting threads. There is no overlap between these actions.
Naming is mixed: 'chat' and 'listmodels' are single compound words, while 'dump_threads' uses an underscore. Using 'list_models' instead of 'listmodels' would improve consistency.
Three tools is slightly on the lean side, but it covers the core functionality of an AI gateway reasonably well. Not too sparse, and each tool earns its place.
The surface covers chat, model listing, and thread export, but lacks a way to list threads without knowing their IDs, and no delete or reset functionality. These are notable gaps for thread management.