Skip to main content
Glama

mcp-reasoner

by Jacck
types.ts1.38 kB
export interface ThoughtNode { id: string; thought: string; score: number; depth: number; children: string[]; // Store child IDs parentId?: string; // Store parent ID isComplete: boolean; } export interface ReasoningRequest { thought: string; thoughtNumber: number; totalThoughts: number; nextThoughtNeeded: boolean; parentId?: string; // For branching thoughts strategyType?: string; // Strategy to use for reasoning beamWidth?: number; // Number of top paths to maintain (n-sampling) numSimulations?: number; // Number of MCTS simulations to run } export interface ReasoningResponse { nodeId: string; thought: string; score: number; depth: number; isComplete: boolean; nextThoughtNeeded: boolean; possiblePaths?: number; bestScore?: number; strategyUsed?: string; } export interface ReasoningStats { totalNodes: number; averageScore: number; maxDepth: number; branchingFactor: number; strategyMetrics?: Record<string, any>; } export const CONFIG = { beamWidth: 3, // Keep top 3 paths maxDepth: 5, // Reasonable depth limit minScore: 0.5, // Threshold for path viability temperature: 0.7, // For thought diversity cacheSize: 1000, // LRU cache size defaultStrategy: 'beam_search', // Default reasoning strategy numSimulations: 50 // Default number of MCTS simulations } as const;

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Jacck/mcp-reasoner'

If you have feedback or need assistance with the MCP directory API, please join our Discord server