tokencost-dev
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_model_detailsA | Look up pricing, context window, and capabilities for an LLM model. Uses fuzzy matching so you don't need the exact model key. |
| calculate_estimateA | Estimate the cost for a given number of input and output tokens on a specific model. Supports optional cached_tokens for prompt caching discounts. |
| compare_modelsA | Filter and compare models by provider, minimum context window, or mode. Returns top 5 most cost-effective matches. |
| refresh_pricesA | Force a re-fetch of pricing data from the LiteLLM registry. Use this if you suspect the cached data is stale. |
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 4 tools
Each tool has a distinct, non-overlapping purpose: estimate cost, compare models, get model details, and refresh pricing data. No confusion possible.
All tool names follow a consistent verb_noun pattern (e.g., calculate_estimate, compare_models), making them predictable and easy to understand.
With 4 tools, the set is appropriately scoped for a token cost estimation server. Each tool serves a necessary function without bloat.
The tool set covers the core lifecycle: estimate costs, compare models, retrieve details, and refresh data. Minor gap: no explicit listing of all available models, though get_model_details uses fuzzy matching.