"Tools for slow thinking, step-back reasoning, and contextual memory capabilities" matching MCP servers:

A Model Context Protocol server that provides Retrieval-Augmented Generation capabilities using Contextual AI, enabling AI interfaces like Cursor IDE and Claude Desktop to query domain-specific knowledge with context-aware responses and source citations.

Provides an MCP server that decomposes complex problems into ordered, isolated reasoning contexts, yielding compressed summaries before discarding intermediate steps. Enables sequential or parallel step execution with optional verification for more reliable LLM reasoning.

Enables MCP clients to offload sub-problems to fresh, isolated contexts via sampling, offering plan, think_aside, and run_plan tools for sequential reasoning without cluttering the main conversation with verbose intermediate steps.

A 66-tool reasoning pipeline that intercepts prompts to classify intent, check past mistakes, and generate execution plans, enabling any LLM to think harder and avoid repeating errors.

An MCP server that enables managing multiple lines of thought with features like branch navigation, cross-references between related thoughts, and insight generation from key points.

A Model Context Protocol (MCP) server implementation for the Google Gemini language model. This server allows Claude Desktop users to access the powerful reasoning capabilities of Gemini-2.0-flash-thinking-exp-01-21 model.

An advanced MCP server that implements sophisticated sequential thinking using a coordinated team of specialized AI agents (Planner, Researcher, Analyzer, Critic, Synthesizer) to deeply analyze problems and provide high-quality, structured reasoning.

Enables AI agents to achieve production-ready solutions through iterative refinement and recursive thinking processes. It features token optimization via context compression and session-based tracking to improve problem-solving depth while minimizing cost.

Provides Dify apps with a secure, app-scoped MCP tool catalog and action gateway for authorized FlowFlox operations, enabling intent-driven planning and fixed operational processes without exposing global credentials.

Enables thinking models to extend their reasoning by outsourcing parts of the chain of thought to a non-thinking model via the chat_agent tool, with configurable parameters.

Enables building and evaluating LangGraph agents over a protocol-level MCP server to measure failure rates across four agent topologies.

Enables AI agents to drive real Google Chrome with stealth and captcha handling, using Patchright to avoid bot detection.

An advanced contextual memory server for agent orchestration providing chronological history, semantic vector search via Redis Stack, and temporary scratchpad storage. It also enables context compression through history rollups and facilitates inter-agent communication using a Redis-based Pub/Sub system.

A passive MCP server that exposes a toolbox of executable tools (shell, network, HTTP, AI search, SSH, S3 file operations) to autonomous agents via Streamable HTTP, with strong security features including Docker sandboxing and WAF.

A shared memory and coordination server for multiple AI coding agents, built on the Model Context Protocol (MCP).

Provides MCP tools for persistent, cross-session memory via Cloudflare Agent Memory, enabling agents to store, recall, ingest, summarize, and manage memory namespaces.

Transforms any API into an intelligent MCP agent in seconds, enabling natural language interaction with APIs through the Model Context Protocol.

An MCP (Model Context Protocol) server implementation using HTTP SSE (Server-Sent Events) connections with built-in utility tools including echo, time, calculator, and weather query functionality.

A Python-based MCP server that provides coordinate conversion between latitude/longitude and UTM/TWD97, along with various civil engineering calculation tools for LLM and AI application integration.

An MCP server that exposes 43 domain tools behind 4 meta-tools, applying guardrails such as budgets, rate limits, schema validation, human-in-the-loop confirmation for mutations, PII redaction, and an append-only audit log while reducing context window usage.