LLM Router MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GOOGLE_API_KEY | No | API key for Google Gemini | |
| OPENAI_API_KEY | No | API key for OpenAI GPT-4o | |
| ANTHROPIC_API_KEY | No | API key for Anthropic Claude |
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 |
|---|---|
| route_promptA | Routes a coding prompt to the best LLM (Claude, Gemini, GPT-4o) based on task type. Automatically classifies the task and selects the cheapest/most capable model. |
| plan_workflowC | Always uses Gemini to plan a high-level workflow or architecture for a feature. |
| generate_codeB | Always uses Claude for complex code generation, logic-heavy tasks, or refactoring. |
| implement_featureC | Always uses GPT-4o for feature implementation, test generation, or repetitive coding tasks. |
| clear_contextB | Clears the context cache for a session, starting fresh. |
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 5 tools
route_prompt overlaps heavily with plan_workflow, generate_code, and implement_feature, as those three are just specialized routing tools with predetermined models. Additionally, generate_code and implement_feature are similar enough (complex vs repetitive coding) that an agent may struggle to choose between them.
All tool names follow a consistent verb_noun pattern (route_prompt, plan_workflow, clear_context, generate_code, implement_feature). The naming is predictable and clearly indicates the action and target.
With 5 tools, the server is well-scoped for a specialized LLM routing purpose. Each tool has a distinct name and fits within the expected 3-15 range, making the tool surface easy to grasp.
The core workflow of routing, planning, generating, and implementing is covered, along with a context reset. Minor gaps exist, such as no tool to list available models or customize routing rules, but agents can work around these by using route_prompt for general tasks.