mcp-model-proxy
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HTTPS_PROXY | No | Proxy setting for Node.js as it does not follow system proxy by default | |
| CLAUDE_MCP_MODEL | No | Default model to use. Defaults to claude-sonnet-4-5 | claude-sonnet-4-5 |
| CLAUDE_MCP_API_KEY | Yes | Upstream API Key for the provider | |
| CLAUDE_MCP_BASE_URL | Yes | Claude compatible upstream address. Supports formats like https://example.com, https://example.com/v1, or https://example.com/v1/messages | |
| CLAUDE_MCP_TIMEOUT_MS | No | Upstream request timeout in milliseconds. Default 60000, minimum 1000 | 60000 |
| CLAUDE_MCP_ANTHROPIC_VERSION | No | Claude API version. Defaults to 2023-06-01 | 2023-06-01 |
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 | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| ask_modelC | Call a model via the local MCP server. Sends a prompt to an upstream service compatible with the Claude Messages API and returns the text response. |
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 1 tool
With only one tool, there is no possibility of ambiguity or overlap between tools. The tool 'ask_model' has a clearly distinct and singular purpose, making misselection impossible.
The single tool name 'ask_model' follows a clear verb_noun pattern, and with only one tool, consistency is inherently perfect as there are no other names to compare or deviate from.
A single tool is too few for a server named 'mcp-model-proxy', which suggests a broader scope for model interactions. This minimal set feels thin and likely incomplete for typical proxy functionalities like listing models, managing configurations, or handling different API endpoints.
The tool surface is severely incomplete for a model proxy domain. While 'ask_model' covers a core operation, there are significant gaps such as listing available models, configuring model parameters, handling streaming responses, or managing API keys, which will likely cause agent failures in real-world scenarios.