MCP MeasureMCP Measure

índice observado

Research Agent by Nova (CIVAI)

I do everything related to research and reports

¿Debería usar esto?

Calidad y seguridad

B
Calidad de la descripción
93%
Integridad del esquema
93%
Calidad de los nombres
55%
Riesgo de envenenamiento
100%
Coincidencia de permisos
100%
Cumplimiento del protocolo
100%

Hallazgos (5)

  • LOWTool 'google_search__converse' description lacks action verben google_search__converse
  • LOWTool 'deep_research__conduct-deep-research' doesn't follow camelCase/snake_caseen deep_research__conduct-deep-research
  • LOWTool 'deep_research__conduct-deep-research' name length outside 3-30 rangeen deep_research__conduct-deep-research
  • LOWTool 'deep_research__generate-research-report' doesn't follow camelCase/snake_caseen deep_research__generate-research-report
  • LOWTool 'deep_research__generate-research-report' name length outside 3-30 rangeen deep_research__generate-research-report

Basado en el análisis automatizado de las definiciones de herramientas y el cumplimiento del protocolo.

Costo de contexto

~432Tokens (definiciones de herramientas)
~1022 BTamaño de respuesta típico
Impacto mínimo en la atención (0.34% del contexto de 128k)

Este es el número aproximado de tokens que se consumen cada vez que las herramientas del servidor se cargan en el contexto de un modelo. Los recuentos más altos reducen la atención disponible para otras tareas.

Instalar

Instalación con un clic

Agrega esto a tu archivo `claude_desktop_config.json`:

{
  "mcpServers": {
    "research-agent": {
      "url": "https://nova.civai.co/mcp/agents/research-agent"
    }
  }
}

Puntos de conexión remotos

https://nova.civai.co/mcp/agents/research-agentstreamable-http

Qué puede hacer

Inventario de herramientas

Herramientas (4)

🟢 Solo lectura🟡 Escritura🔴 Eliminación⚪ Desconocido
🟢deep_research__conduct-deep-research(topic, objective, breadth, depth, max_total_queries, ...)

Conduct deep, iterative research on a topic by generating multiple search queries, processing the results, and recursively exploring new research directions.

Esquema de entrada

{
  "type": "object",
  "properties": {
    "topic": {
      "type": "string",
      "description": "The main topic or question to research"
    },
    "objective": {
      "type": "string",
      "description": "The specific goal or objective of the research"
    },
    "breadth": {
      "type": "integer",
      "description": "Number of search queries per research direction (1-5)",
      "default": 3,
      "minimum": 1,
      "maximum": 5
    },
    "depth": {
      "type": "integer",
      "description": "Depth of recursive exploration (0-2)",
      "default": 1,
      "minimum": 0,
      "maximum": 2
    },
    "max_total_queries": {
      "type": "integer",
      "description": "Maximum number of search queries to process (2-5)",
      "default": 2,
      "minimum": 2,
      "maximum": 5
    },
    "max_duration_seconds": {
      "type": "integer",
      "description": "Maximum duration for the research process in seconds (60-300)",
      "default": 120,
      "minimum": 60,
      "maximum": 300
    },
    "connection_id": {
      "type": "string",
      "description": "Optional connection ID for the user",
      "default": ""
    }
  },
  "required": [
    "topic",
    "objective"
  ]
}
⚪deep_research__generate-research-report(research_data, objective)

Generate a comprehensive research report from existing research data.

Esquema de entrada

{
  "type": "object",
  "properties": {
    "research_data": {
      "type": "string",
      "description": "The research data and findings to include in the report"
    },
    "objective": {
      "type": "string",
      "description": "The specific goal or objective of the report"
    }
  },
  "required": [
    "research_data",
    "objective"
  ]
}
🟢google_search__search(query)

Run live Google search and summarize results.

Esquema de entrada

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string"
    }
  },
  "required": [
    "query"
  ]
}
🟢google_search__converse(reply_hint)

Reply conversationally when the request is ambiguous or needs clarification.

Esquema de entrada

{
  "type": "object",
  "properties": {
    "reply_hint": {
      "type": "string",
      "description": "Optional hint for the conversational reply."
    }
  }
}

Prompts recomendados

search_research
Search for information about [topic] using Research Agent by Nova (CIVAI)
Herramientas esperadas: deep_research__conduct-deep-research
find_specific
Find [specific item] using Research Agent by Nova (CIVAI)
Herramientas esperadas: deep_research__conduct-deep-research

Comunidad

Califica este servidor

Evidencia

Observaciones recientes

verificadoversión no registrada4 herramientas
verificadoversión no registrada4 herramientas