Train in Silence
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| planner_metadataA | Return MCP server metadata and the exposed planner capabilities. |
| validate_requestB | Validate a planning request provided inline or loaded from a YAML/JSON config path. |
| recommend_hardwareC | Generate ranked hardware recommendations for an LLM fine-tuning planning request. |
| explain_planA | Run the full planner and return estimates, normalized market offers, and provider statuses. WARNING: This tool returns raw market data and is extremely verbose (often 100k+ tokens). Use recommend_hardware for standard recommendation tasks. |
| list_providersC | Fetch provider health statuses using optional planning constraints. |
| probe_marketC | Run market aggregation for a planning request and return provider statuses. |
| dump_market_offersC | Return normalized market offers considered for a planning request. |
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 7 tools
Some tools overlap in purpose (e.g., explain_plan and recommend_hardware both generate recommendations, but one is warned against using), and dump_market_offers vs probe_market vs list_providers have similar market data focus, causing potential confusion.
Most tools use snake_case with verb-first (dump_, explain_, list_, probe_, recommend_, validate_), but planner_metadata breaks the pattern with a noun-first name.
With 7 tools, the server covers core functionalities for planning and recommendation without being excessive or too sparse.
The tool set covers market queries, validation, and recommendations, but lacks tools for modifying or managing plans or configurations, leaving minor gaps.