◈──◆──◇ GIKENDAASOWIN AABAJICHIGANAN MCP SERVER / COGNITIVE TOOLS MCP SERVER ◇──◆──◈

What is ◈──◆──◇ GIKENDAASOWIN AABAJICHIGANAN MCP SERVER / COGNITIVE TOOLS MCP SERVER ◇──◆──◈?

An MCP server that provides the deliberate tool implementing an LLM-guided Enhanced 2-Round, 6-Stage Cognitive Deliberation Framework. It prompts LLMs to evaluate and select from 15 modern cognitive techniques using a 0.00-0.99 scoring system with a ≥1.53 threshold rule, accepting only input and context parameters.

How to use ◈──◆──◇ GIKENDAASOWIN AABAJICHIGANAN MCP SERVER / COGNITIVE TOOLS MCP SERVER ◇──◆──◈?

Install via npm using either @nbiish/gikendaasowin-aabajichiganan-mcp or @nbiish/cognitive-tools-mcp. Add it as an MCP server to Claude Desktop by editing claude_desktop_config.json with the npx command and the chosen package name. Then invoke the deliberate tool with the required input parameter and optional context.

Key features of ◈──◆──◇ GIKENDAASOWIN AABAJICHIGANAN MCP SERVER / COGNITIVE TOOLS MCP SERVER ◇──◆──◈

• LLM-guided dynamic technique selection
• 15 modern cognitive prompting strategies
• 0.00-0.99 scoring with ≥1.53 threshold
• Meta-cognitive enhancement for adaptive reasoning
• Built-in iterative improvement guidance
• Fix for "ReferenceError: prompt is not defined"

Use cases of ◈──◆──◇ GIKENDAASOWIN AABAJICHIGANAN MCP SERVER / COGNITIVE TOOLS MCP SERVER ◇──◆──◈

• LLM self-selecting optimal reasoning approaches for complex problems
• Enhancing prompt engineering with structured cognitive frameworks
• Applying adaptive strategy combinations to context-specific tasks
• Iteratively refining LLM outputs through re-deliberation cycles

FAQ from ◈──◆──◇ GIKENDAASOWIN AABAJICHIGANAN MCP SERVER / COGNITIVE TOOLS MCP SERVER ◇──◆──◈

How does this server differ from static prompt templates?

Unlike fixed templates, the deliberate tool prompts the LLM to evaluate and score techniques dynamically, enabling truly adaptive reasoning without hardcoded evaluations.

What are the runtime dependencies?

The server requires Node.js and npm. It is built as an MCP server using TypeScript and runs via the npx command.

Where does the data I input go?

The README does not specify data storage; the tool processes the input and context parameters within the LLM session and returns results.

Are there any known limits or bugs?

A critical bug in v10.0.2 has been fixed: the "ReferenceError: prompt is not defined" error that prevented LLM usage is resolved, ensuring reliable operation.

What transport or authentication does this server use?

The server is an MCP server for Claude Desktop; no authentication or transport details beyond the standard MCP protocol are mentioned.