jev-netlify-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| JEV_PROXY_KEY | No | Your proxy key | |
| JEV_PROXY_URL | No | Your Netlify site address (the key-locked proxy URL) | |
| TYPESAFE_API_KEY | No | Fallback API key so the server also works with the TypeSafe API directly | |
| TYPESAFE_BASE_URL | No | Fallback base URL so the server also works with the TypeSafe API directly | |
| JEV_NETLIFY_MCP_NO_REGISTRY | No | Set to 1 to skip the user registry environment lookup |
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 |
|---|---|
| jev_evaluateB | Evaluate one text (state) against typed questions with Jev and return calibrated answers. state: string or JSON object (English works best). questions: {name: {type: 'noul'|'choice'|'score', instructions: str, criteria?: ...}}. noul criteria optional {true, false}; choice criteria {option: description}; score criteria [level0, level1, ...] from low to high. model defaults to jev-latest. |
| jev_modelsB | List the Jev models and aliases available on this endpoint. |
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 2 tools
The two tools are clearly distinct: one evaluates text against typed questions, while the other lists available models. There is no overlap in purpose or parameters that would cause misselection.
Both names use a consistent jev_ prefix and snake_case, but jev_evaluate is action-oriented while jev_models is a resource listing. This is a minor deviation from a pure verb_noun pattern.
Two tools is thin for a server, though each tool has a valid role in a minimal evaluation API wrapper. It is borderline rather than clearly well-scoped or severely mismatched.
The core evaluation workflow and model discovery are covered, but there are minor gaps such as no batch evaluation, validation, or result-retrieval tooling. These are workable omissions for a thin MCP wrapper.