jev-norma-mcp

triage

Normalize Spanish-language data-privacy policies or terms into clauses and classify each as obligation, risk, or no-action with calibrated confidence, routing low-confidence clauses to human review.

Instructions

Triaje normativo LatAm (gate System One). Toma un documento legal (HTML o PDF de una política de tratamiento de datos / términos en español, Ley 1581 Colombia) o texto/HTML inline, normaliza cláusulas (heurística 0-IA sin palabras), y clasifica cada cláusula con Jev en choice [obligacion|riesgo|sin_accion] + confidence calibrada. Devuelve un reporte JSON con cola de revisión humana: toda cláusula con confianza < 0.8 marca gate_humano=true (fail-open: nunca matar por Jev solo).

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "triage"
  }
}

Input Schema

Name	Required	Description
`api`	No	Si false, solo normaliza (sin llamar a la API de Jev). Default true.
`text`	No	Texto o HTML inline (si no se pasa 'source').
`type`	No	Tipo del contenido. 'pdf' solo tiene sentido con source. Por defecto se infiere.
`sample`	No	Cláusulas a enviar al gate Jev (0 = todas). Default 12.
`source`	No	Ruta a un archivo .html/.pdf. Alternativo a 'text'.

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Oct 28, 2026

TDQS

A4.1/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the behavioral burden and does well: it discloses heuristic normalization, Jev classification into a fixed choice set, calibrated confidence, a human gate at confidence < 0.8, and fail-open behavior. It does not cover permissions, limits, or full operational failure modes, but the key non-obvious behavior is present.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a compact paragraph that front-loads the domain and then explains processing and output behavior. It is dense but not padded; minor parenthetical abbreviations add some reading friction without costing much.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given five parameters, no annotations, and no output schema, the description explains enough to select and invoke the tool: input forms, classification categories, confidence threshold, human-review gate, and fail-open semantics. It does not fully specify the JSON report shape or the role of every parameter, but the schema fills those gaps.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents all five parameters and defaults. The description adds domain-level input context (HTML/PDF policy document or inline text/HTML), but does not explain parameter-specific behavior such as sample size, api=false mode, or type inference.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific operation and domain: it triages LatAm legal data-treatment terms (Ley 1581 Colombia), normalizes clauses, and classifies each clause as obligacion/riesgo/sin_accion with confidence. It also states the return artifact, a JSON report with a human-review queue, so the agent can tell exactly what the tool produces.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives clear usage context: legal policy terms in Spanish, HTML/PDF source documents, or inline text/HTML. There are no sibling tools to route between, but it also does not state explicit exclusions or when not to use the tool.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools

triageA