jev-legal-clause-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| JEV_MODEL | No | Model name | jev-latest |
| JEV_BASE_URL | No | Override for self-hosted / proxy | https://api.typesafe.ai/v1/systemone |
| JEV_MCP_MOCK | No | 1 forces mock | 0 |
| TYPESAFE_API_KEY | No | Required for real calls; blank falls back to mock | |
| JEV_MCP_TIMEOUT_MS | No | Per-request timeout | 30000 |
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 |
|---|---|
| flag_clausesA | Legal-risk clause flagger powered by Jev (System One decision model). Given a contract excerpt, returns calibrated presence probabilities for 10 default risk categories (indemnification / non-compete / confidentiality / termination penalty / auto-renewal / liability cap / IP assignment / governing law / payment terms / personal guarantee) plus an overall risk score and a review recommendation. This is a triage aid — NOT a legal opinion. Always have a human review before signing. |
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
With only a single tool, there is no possibility of confusing it with another tool. The purpose is unambiguous and clearly described.
The tool name flag_clauses follows a clear verb_noun convention. Even with only one tool, the name is descriptive, consistent, and easy to predict.
A single tool feels extremely thin for an MCP server. While specialized, the server would benefit from additional tools such as batch processing, clause list retrieval, or analysis history to justify its existence as a server.
The core flagging operation is present and functional, but the surface is minimal. There are notable gaps such as no way to retrieve past analyses, no configuration of risk categories, and no multi-clause or document-level handling, which limits agent workflows.