Skip to main content
Glama
enesjakozh
Jacck

mcp-reasoner

by Jacck
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

Razonador MCP

Una implementación de razonamiento para Claude Desktop que le permite usar tanto Beam Search como Monte Carlo Tree Search (MCTS). Sinceramente, esto comenzó como una forma de ver si podíamos hacer que Claude fuera aún mejor en la resolución de problemas complejos... resulta que definitivamente podemos.

Versión actual:

versión 2.0.0

Qué hay de nuevo:

Se agregaron 2 algoritmos de razonamiento experimental:

- `mcts-002-alpha`

    - Uses the A* Search Method along with an early *alpha* implementation of a Policy Simulation Layer

    - Also includes an early *alpha* implementation of Adaptive Exploration Simulator & Outcome Based Reasoning Simulator

*NOTE* the implementation of these alpha simulators is not complete and is subject to change

- `mcts-002alt-alpha`

    - Uses the Bidirectional Search Method along with an early *alpha* implementation of a Policy Simulation Layer

    - Also includes an early *alpha* implementation of Adaptive Exploration Simulator & Outcome Based Reasoning Simulator

*NOTE* the implementation of these alpha simulators is not complete and is subject to change

¿Qué pasó con mcts-001-alpha y mcts-001alt-alpha ?

En pocas palabras: era inútil y prácticamente similar al método mcts básico. Tras las pruebas iniciales, los resultados obtenidos en los procesos de pensamiento básicos fueron prácticamente similares, lo que demuestra que añadir simplemente simulación de políticas podría no tener ningún efecto.

Entonces, ¿por qué agregar la capa de simulación de políticas ahora?

Bueno, creo que es importante incorporar políticas y búsqueda en conjunto, ya que así es como la mayoría de los algoritmos las implementan.

Versiones anteriores:

versión 1.1.0

Se agregó control del modelo sobre los parámetros de búsqueda:

beamWidth: permite a Claude ajustar la cantidad de rutas a rastrear (1-10)

numSimulations: ajuste preciso del recuento de simulaciones MCTS (1-150)

Características

  • Dos estrategias de búsqueda entre las que puedes alternar:

    • Búsqueda de haz (útil para cosas sencillas)

    • MCTS (cuando las cosas se ponen complejas) con variaciones alfa (ver arriba)

  • Realiza un seguimiento de la calidad de los diferentes caminos de razonamiento.

  • Describe todas las diferentes formas en las que Claude piensa en los problemas.

  • Analiza cómo fue el proceso de razonamiento.

  • Sigue el protocolo MCP (obviamente)

Related MCP server: MCP Advanced Reasoning Server

Instalación

git clone https://github.com/frgmt0/mcp-reasoner.git 

OR clone the original:

git clone https://github.com/Jacck/mcp-reasoner.git

cd mcp-reasoner
npm install
npm run build

Configuración

Agregar a la configuración de Claude Desktop:

{
  "mcpServers": {
    "mcp-reasoner": {
      "command": "node",
      "args": ["path/to/mcp-reasoner/dist/index.js"],
    }
  }
}

Pruebas

[Más pruebas próximamente]

Puntos de referencia

[Se añadirá próximamente la evaluación comparativa]

Puntos de referencia clave para realizar pruebas:

  • MATH500

  • GPQA-Diamante

  • GMSK8

  • Quizás Polyglot y/o SWE-Bench

Licencia

Este proyecto está licenciado bajo la licencia MIT: consulte el archivo de LICENCIA para obtener más detalles.

Available Tools

1 tool
mcp-reasonerC

Advanced reasoning tool with multiple strategies including Beam Search and Monte Carlo Tree Search

ParametersJSON Schema
NameRequiredDescriptionDefault
beamWidthNoNumber of top paths to maintain (n-sampling). Defaults to 3 if not specified
nextThoughtNeededYesWhether another step is needed
numSimulationsNoNumber of MCTS simulations to run. Defaults to 50 if not specified
strategyTypeNoReasoning strategy to use (beam_search or mcts)
thoughtYesCurrent reasoning step
thoughtNumberYesCurrent step number
totalThoughtsYesTotal expected steps

TDQS

C2.9/5.0
Behavior2/5

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

No annotations are provided, so the description carries the full burden. It does not disclose behavioral traits such as rate limits, idempotency, or side effects. The description only mentions 'advanced reasoning' without explaining what operations are performed or their implications.

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 single sentence, highly concise. It front-loads the core purpose. However, it may be too brief for the tool's complexity, but it is not verbose.

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

Completeness2/5

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

Given the complexity (7 parameters, no output schema, no annotations), the description is insufficient. It does not explain the reasoning process, how strategies differ, or how the parameters like beamWidth and numSimulations interact. More context is needed for effective use.

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?

The input schema has 100% coverage with descriptions for all 7 parameters. The description adds no additional meaning beyond what the schema already provides. Baseline score of 3 is appropriate as the schema handles the semantics well.

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

Purpose4/5

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

The description clearly states the tool is for advanced reasoning with specific strategies (Beam Search, MCTS). It is specific about the resource and action, though not as precise as a verb+resource pairing. No siblings exist to differentiate from, so the description adequately distinguishes itself.

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

Usage Guidelines2/5

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

The description provides no guidance on when to use this tool versus alternatives, no context on strategy selection, and no prerequisites or use cases. The sibling-tools list is empty, but the description should still offer usage suggestions.

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

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 1 tool updatev1.0.0
    • First observedmcp-reasoner
Deploy Server
Context All Your Agents Can ShareAgents share what they learn, hand off work, and hold it when you switch models. Start with 5 GB free.fulcradynamics.comAd

TDQS

B3/5.0

Scored across 1 tool

Disambiguation5/5

With only one tool, there is no possibility of ambiguity or overlap between tools. The single tool 'mcp-reasoner' has a clearly distinct purpose focused on advanced reasoning strategies.

Naming Consistency5/5

A single tool inherently has perfect naming consistency, as there are no other tools to compare it against. The name 'mcp-reasoner' follows a clear, descriptive pattern that matches the server name.

Tool Count2/5

A single tool is generally too few for most server purposes, as it limits functionality and flexibility. While it might be appropriate for a highly specialized niche, the description suggests advanced reasoning with multiple strategies, which could benefit from more granular tools (e.g., separate tools for different search strategies).

Completeness3/5

It is difficult to assess completeness with only one tool, as there is no clear domain surface to evaluate. The tool covers reasoning strategies, but without additional tools (e.g., for input/output, configuration, or specific reasoning modes), there are likely gaps in a full reasoning workflow.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

Appeared in Searches