deliberate-reasoning-engine and crash-mcp
About deliberate-reasoning-engine
haasonsaas/deliberate-reasoning-engine
MCP server that transforms linear AI reasoning into structured, auditable thought graphs
Implements semantic thought categorization (objectives, hypotheses, assumptions, evidence, critiques) as nodes in a directed acyclic graph with explicit dependency tracking and cascade invalidation—when an assumption is disproven, all dependent reasoning automatically becomes stale. Integrates with Claude Desktop via the Model Context Protocol, exposing tools for logging structured thoughts, retrieving the reasoning graph, and invalidating assumptions to maintain reasoning consistency.
About crash-mcp
nikkoxgonzales/crash-mcp
MCP server for structured and efficient reasoning with step validation, branching, and revisions.
Implements step-by-step reasoning as an MCP server with confidence scoring, revision tracking, and multi-branch exploration capabilities integrated with Claude, Cursor, VS Code, and other AI development environments. Uses stdio transport with automatic session management, supports strict and flexible validation modes, and offers multiple output formats (console, JSON, Markdown) with configurable history limits and branch depth constraints. More token-efficient than sequential thinking through streamlined prompting while maintaining dependency validation for logical coherence across reasoning chains.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work