Levelang MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MCP_PORT | No | Port when using HTTP transport | 8080 |
| MCP_TRANSPORT | No | Transport: stdio or streamable-http | stdio |
| LEVELANG_API_KEY | No | Service key (sk_xxx) for backend auth. Required when connecting to a remote backend. | |
| LEVELANG_API_BASE_URL | No | Levelang backend URL | http://localhost:8000/api/v1 |
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 |
|---|---|
| translateA | Translate text to a target language at a specific proficiency level. Unlike standard translators that produce native-speaker complexity, this translates at the learner's level -- beginner translations use simple grammar, intermediate uses more complex structures, etc. Args: text: The text to translate (any length, any source language) target_language: Target language code -- use list_languages to see available codes (e.g. fra, deu, cmn, yue, ita) level: Proficiency level -- proficiency levels available for the target language (e.g. beginner, intermediate, advanced, and/or fluent) source_language: Source language code (default: eng for English) mood: Tone -- tones available for the target language mode: Language mode (spoken/written) -- controls whether the translation targets written or spoken register. Use list_languages to see available modes per language. Returns: The translated text with metadata about the translation. |
| list_languagesA | List all languages supported by Levelang with their available levels and moods. Use this to discover valid language codes, proficiency levels, mood options, and mode options before calling the translate tool. Returns: Formatted list of supported languages and their configurations. |
| translate_compareA | Translate text at multiple proficiency levels to compare complexity differences. Shows how the same text is translated differently at different levels -- useful for understanding how grammar and vocabulary constraints change across proficiency. Args: text: The text to translate (any length, any source language) target_language: Target language code -- use list_languages to see available codes (e.g. fra, deu, cmn, yue, ita) source_language: Source language code (default: eng for English) mood: Tone -- tones available for the target language levels: Optional list of proficiency level codes to compare (e.g. ["beginner", "advanced"]). If omitted, compares all available levels. Use list_languages to see valid codes per language. mode: Optional language mode (spoken/written) -- controls whether the translation targets written or spoken register. If omitted, compares all available levels. Use list_languages to see valid codes per language. Returns: The same text translated at each requested level, formatted for comparison. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| compare_levels | Create a prompt for comparing how the same text is translated at different levels. Args: language: Target language name |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| languages_resource | List of all supported languages with their levels and moods. |
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: list_languages provides configuration discovery, translate performs single-level translation, and translate_compare enables multi-level comparison. There is no overlap in functionality, and an agent can easily distinguish between them based on their specific objectives.
All tool names follow a consistent verb_noun pattern: list_languages, translate, and translate_compare. The naming is predictable and readable, with no deviations or mixed conventions, making it easy for agents to understand the tool set's structure.
With 3 tools, the count is reasonable for a language translation server, covering core operations like discovery, translation, and comparison. However, it feels slightly thin, as additional tools for managing translations or handling errors could enhance completeness, but the current set is well-scoped for basic functionality.
The tool set covers essential workflows: discovering supported languages, translating text at specific levels, and comparing translations across levels. Minor gaps exist, such as no tools for updating or deleting translations, but these are not critical for the server's purpose, and agents can work effectively with the provided operations.