What safety limits does the server have?
The server automatically stops after 20 thought steps to prevent infinite loops.
@mettamatt
Basic information
A code reasoning MCP server, a fork of sequential-thinking
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"code-reasoning": {
"command": "npx",
"args": [
"-y",
"@mettamatt/code-reasoning"
]
}
}
}π§ A detailed tool for dynamic and reflective problem-solving through sequential thinking. This tool helps you analyze problems through a flexible thinking process that can adapt and evolve. Each thought can build on, question, or revise previous insights as understanding deepens. π KEY PARAMETERS: - thought: Your current reasoning step (required) - thought_number: Current number in sequence (required) - total_thoughts: Estimated final count (required, can adjust as needed) - next_thought_needed: Set to FALSE ONLY when completely done (required) - branch_from_thought + branch_id: When exploring alternative approaches (πΏ) - is_revision + revises_thought: When correcting earlier thinking (π) β CRITICAL CHECKLIST (review every 3 thoughts): 1. Need to explore alternatives? β Use BRANCH (πΏ) with branch_from_thought + branch_id 2. Need to correct earlier thinking? β Use REVISION (π) with is_revision + revises_thought 3. Scope changed? β Adjust total_thoughts up or down as needed 4. Only set next_thought_needed = false when you have a complete, verified solution π‘ BEST PRACTICES: - Start with an initial estimate of total_thoughts, but adjust as you go - Don't hesitate to revise earlier conclusions when new insights emerge - Use branching to explore multiple approaches to the same problem - Express uncertainty when present - Ignore information that is irrelevant to the current step - End with a clear, validated conclusion before setting next_thought_needed = false βοΈ End each thought by asking: "What am I missing or need to reconsider?"
A Model Context Protocol (MCP) server that enhances Claude's ability to solve complex programming tasks through structured, step-by-step thinking. It integrates with Claude Desktop and VS Code, providing a systematic approach to coding challenges.
Install by adding the server configuration (npx -y @mettamatt/code-reasoning) to your Claude Desktop or VS Code MCP settings. To trigger reasoning, append "Use sequential thinking to reason about this." to your chat message, or use ready-made prompt templates via the "+" icon or /help command. Optional command-line flags include --debug for detailed logging and --help for help.
The server automatically stops after 20 thought steps to prevent infinite loops.
You can use `--debug` to enable detailed logging and `--help` (or `-h`) to show help information.
Append "Use sequential thinking to reason about this." to your chat message, or select a ready-to-use prompt from the Code Reasoning prompts menu.
The server works with Claude Desktop and VS Code via standard MCP configuration.
The project is licensed under the MIT License.
ArduPilotγγγΌγ³γAIγ¨γΌγΈγ§γ³γγγζδ½γγMCPγ΅γΌγγΌγ§γγ
An MCP (Model Context Protocol) server that helps AI assistants (like Claude) break down complex tasks into manageable steps, track progress, and manage a hierarchical task list.
Neuronto Agentic Resource Discovery (ARD) Index. Federated search across every public ARD registry, plus a verified tool index read from each MCP server's own tools/list, hybrid lexical and semantic retrieval, and ARD-Bench.
Agentic SWMM is an automated, auditable, and memory-informed framework for reproducible stormwater modelling, integrating QGIS and EPA SWMM through the aiswmm runtime, reusable Skills, and MCP interfaces, with QA verification, provenance tracking, calibration support, and Codex,
MCP server that allow LLM in agent mode builds n8n workflows for you
Comments