Jev MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| JEV_MODEL | No | Optional. The model identifier to use for evaluations. Defaults to 'jev-1.13.0'. Set to 'jev-latest' or another specific version to override. | |
| TYPESAFE_API_KEY | Yes | Required. Your TypeSafe API key, used to authenticate requests to the TypeSafe API. |
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 |
|---|---|
| jev_evaluateA | Evaluate supplied text or JSON with TypeSafe Jev: classify records, triage tickets, score quality, judge relevance, compare options, or audit content. Mix independent noul (probability of yes), choice (named options), and score (ordered rubric) questions in one call. Returns typed answers, probabilities, confidence where available, model and token usage. For batches, include all items in state and explicitly identify each target item in its question instructions; question IDs are not seen by the model. Does not fetch URLs, browse, generate explanations, or modify data. Sends supplied content to the external TypeSafe API and consumes quota. Review uncertain or consequential results. |
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
There is only one tool, so an agent cannot confuse it with another tool. Its name and description clearly identify it as the evaluation entry point.
The sole tool uses a clear snake_case verb_noun convention ('jev_evaluate'), consistent with a single-tool surface. No conflicting naming patterns exist.
One tool is appropriate because the server is a focused wrapper around a single Jev evaluation operation. The tool is broad enough to handle classification, scoring, choice, and batch use cases, so additional tools would add little value.
For its stated scope—evaluating supplied text/JSON—the tool covers all core needs: multiple question types, batch evaluation, typed answers, confidence, and usage metadata. The explicit exclusions (fetching URLs, generating explanations) are outside the server's purpose, not missing features.