MCP.so
ログイン

Deep Thinker

@hubinoretros

リポジトリ

Deep Thinker について

Advanced cognitive thinking MCP server with DAG-based thought graph, multiple reasoning strategies, metacognition, and self-evaluation.

#mcp-server#thinking-tool#reasoning-engine#cognitive-engine

基本情報

カテゴリ

その他

トランスポート

stdio

公開者

hubinoretros

投稿者

Hubino HOTEL

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "deep-thinker": {
      "command": "npx",
      "args": [
        "-y",
        "deep-thinker"
      ]
    }
  }
}

ツール

6

Add a thought to the cognitive graph using the current strategy. Supports sequential, dialectic, parallel, analogical, and abductive reasoning strategies. Each thought becomes a node in a DAG with confidence scoring, edges, and metadata.

Evaluate the thinking process: score confidence, generate critiques, and assess overall graph health. Provides detailed analysis of weak spots and strong reasoning paths.

Metacognitive operations: view the current thinking state, get strategy suggestions, switch strategies, and receive guidance on improving reasoning. The system automatically detects stuck states and recommends actions.

Query and visualize the thought graph. View the DAG structure, find paths, inspect branches, and get statistics.

Prune and optimize the thought graph. Remove dead ends, consolidate redundant branches, and optimize reasoning paths. Helps maintain graph efficiency during deep reasoning.

Reset the thought graph and metacognitive state. Start a fresh reasoning session.

概要

What is Deep Thinker?

Deep Thinker is an advanced cognitive MCP server with DAG-based thought graphs, multiple reasoning strategies, metacognition, and self-evaluation. It extends sequential thinking by supporting branching, merging, and cross-edges.

How to use Deep Thinker?

—

Key features of Deep Thinker

  • DAG-based thought graph with branching, merging, and cross-edges
  • Five reasoning strategies: Sequential, Dialectic, Parallel, Analogical, Abductive
  • Multi-factor confidence scoring with support/contradiction analysis
  • Automatic self-critique with severity levels and confidence adjustments
  • Metacognitive engine detecting stuck states and stagnant reasoning
  • Knowledge integration for attaching, gap detection, and consistency validation
  • Thought pruning: dead-end detection, redundancy removal, branch elimination

Use cases of Deep Thinker

  • Complex problem solving that requires structured deep reasoning
  • Scenarios needing dialectical or abductive inference for best explanations
  • Tasks where self-evaluation and confidence calibration are critical
  • Exploratory analysis with branching and merging of multiple reasoning paths
  • Knowledge-intensive reasoning requiring external knowledge integration and validation

FAQ from Deep Thinker

What reasoning strategies does Deep Thinker support?

Deep Thinker supports Sequential, Dialectic (thesis→antithesis→synthesis), Parallel, Analogical, and Abductive (inference to best explanation) strategies.

How does the metacognitive engine work?

The metacognitive engine detects stuck states, stagnation, and declining confidence, then suggests strategy switches and corrective actions.

What is thought pruning and why is it needed?

Thought pruning removes dead-ends, redundant thoughts, and deep unproductive branches to optimize the reasoning graph.

How is confidence scoring calculated?

Confidence scoring uses multi-factor evaluation including support/contradiction analysis, depth penalties, and knowledge integration boosts.

How does Deep Thinker differ from sequential-thinking MCP?

Deep Thinker uses a DAG-based thought graph with branching, merging, and cross-edges rather than a linear chain, and adds multiple reasoning strategies, confidence scoring, self-critique, metacognition, knowledge integration, and pruning.

コメント

「その他」の他のコンテンツ

Maestro

mobile-dev-inc

Painless E2E Automation for Mobile and Web

14.8K

IDA Pro MCP

mrexodia

AI-powered reverse engineering assistant that bridges IDA Pro with language models through MCP.

9.9K

Inbox Zero AI

elie222

The world's best AI personal assistant for email. Open source app to help you reach inbox zero fast.

11.6K

Unity MCP ✨

justinpbarnett

Unity MCP acts as a bridge between AI assistants and your Unity Editor. Give your LLM tools to manage assets, control scenes, edit scripts, and automate tasks within Unity.

12.2K

Reactive Resume

amruthpillai

A one-of-a-kind resume builder that keeps your privacy in mind. Completely secure, customizable, portable, open-source and free forever. Try it out today!

39.5K