MCP Reasoner

MCPOpen SourceMIT63.7

by Jacck • Search & Retrieval

A reasoning server that provides Beam Search and MCTS-based reasoning (with experimental alpha variants) for Claude Desktop to improve complex problem-solving.

Example Use Cases

1
Perform structured multi-step reasoning using Beam Search for straightforward tasks.
2
Explore complex decision spaces using MCTS with tunable simulation counts and experimental policy-simulation layers.
3
Log, analyze, and compare alternative reasoning paths and outcomes for diagnostics or research.

Description

MCP Reasoner implements two search strategies—Beam Search and Monte Carlo Tree Search—with experimental alpha variations that add policy simulation and adaptive exploration. It tracks and scores multiple reasoning paths, maps out how decisions were reached, and analyzes the reasoning process following the MCP protocol. The project is designed to integrate into Claude Desktop as an external MCP server, with configurable search parameters like beamWidth and numSimulations to tune performance.

Quick Actions

View on GitHub

Security

Scanned 3 month(s) ago

Risk Level

MINIMAL

Read-only data retrieval, no side effects

Trust Score

D44/100

4/17 checks passed

Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.

Quick Stats

Service TypeMCP

Pricing ModelFree

Capabilities0 Tools / 0 Prompts / 0 Resources

OwnerJacck

CategorySearch & Retrieval

DependenciesStandalone