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

Persistent memory for AI coding agents: local-first semantic search, dependency graphs, git history.

Enables MCP clients to perform step-level process reward evaluation, MCTS-guided reasoning search, hallucination branch pruning, and formal verification for complex math, code, and logic problems.

Persistent sequential thinking for MCP clients, stored durably on Cloudflare Workers with support for branching, revision, and history retrieval.

An enhanced sequential thinking tool optimized for programming tasks that helps break down complex coding problems into structured, self-auditing thought steps with branching and revision capabilities.

MCP server for structured reasoning with cognitive trap detection, verification, and context compression

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.

Cross-session memory for AI coding agents: six tools (save/search/list/get/update/delete) over a local-first SQLite store with human-readable Markdown mirror, semantic search, background consolidation, and human-reviewed conflict resolution. Requires the dsh-mneme DSH plugin running locally.

Search and pull 620+ React and Tailwind sections into your project over MCP.

Enables AI agents and MCP clients to compress retrieved web documents against a query, pruning irrelevant noise through deterministic algorithmic scoring so downstream context stays lean. It exposes this capability over the Model Context Protocol for integration with Claude Desktop, Cursor, and similar clients.

This MCP server enables clients to reduce oversized prompts by selecting task-relevant context, removing lexical duplicates, and pruning to a token budget while preserving critical system invariants.

Enables deterministic calculations for AI API and subscription costs, meeting time zones, recurring schedules, and cron next-run previews, returning normalized inputs, units, assumptions, dated sources, and links to results.

Educational tools agent by Birkan Yavuz — an A2A + MCP server for classroom tools, with creator attribution in every result. It offers a catalog of 24 real classroom activities, including math visualizers and reading and sight-word tools.

A local MCP server that gives AI assistants persistent long-term memory using a biological neural architecture with cortex, synapses, and hippocampus.

Local-first personal memory for AI agents. Ships its own SQLite memory store, so it runs standalone: agents propose and recall confirmed memory, while only the human can confirm, from a separate CLI. Optionally uses the macOS Hive app for encrypted vaults. Also provides a shared board for coordinating several agents.

MCP server that exposes agent-memory-daemon to any MCP-compatible client — Kiro (CLI & IDE), Claude Desktop, Cursor, and others. The daemon does the thinking (consolidation + extraction); this server is a thin filesystem bridge so agents can read, append, and search memory through the Model Context Protocol.

Long-term memory for AI coding agents, stored as a tree of plain local files. Agents search it before answering and save what lasts; changed facts are kept as history and flagged as outdated.

Provides persistent local memory for any MCP-compatible AI agent, enabling agents to store, search, recall, forget, export, and manage preferences, rules, and facts with zero dependencies, no cloud, and no API keys.

A different approach from typical persistent-memory MCPs. Instead of a local SQLite + embeddings store, the memory lives as plain files in a .ai-memory/ directory you commit to your repo (facts.jsonl, decisions/\*.md, gotchas.md). Git is the sync layer — what one Claude/Cursor/Cline learns about a repo, the next session (or a teammate's agent) picks up automatically. 5 MCP tools: get_rep

MARM MCP provides persistent memory and structured session context beneath any AI tool, so your agents learn, remember, and collaborate across all your workflows.

Hand a project over — to your next session, to Cursor/Codex/Gemini, to your successor — with the reasons attached. Local-first memory MCP for coding agents: decisions re-injected before the agent acts, drift detection across every project you run. One SQLite file, nothing leaves your machine.

A stateless MCP server that gives coding agents a curated entry point to shared memory, orchestrating Hindsight (facts) and LightRAG (corpora) for automatic routing, URL ingestion, and consistency tooling.

Local-first persistent memory MCP: shared SQLite key/value store, searchable, TTL-aware and secret-safe.

A portable, self-hosted semantic memory layer for LLM agents, enabling them to search and add memories by meaning over the Model Context Protocol.

Two-layer memory for AI agents. Episodes compress into identity. The only MCP memory server with an immune system. Patterns earn permanence through evidence, false knowledge gets caught and demoted, and stale information fades — so your agent's memory gets smarter over time, not just bigger. Zero dependencies. 5 tools. Works with any MCP client.

Provides AI agents with persistent, long-term memory via OKF-formatted markdown and SQLite indexing, enabling stateful storage, retrieval, and search across sessions.

Persistent memory and automatic git snapshots for Claude Code, capturing decisions, patterns, and architecture across sessions.

Persistent project memory for coding agents, kept in a local SQLite store with FTS5 search: search and list memories by topic, read one with its history, add a fact, show the store's statistics, and read or regenerate the PROGRESS.md handoff. Pure standard-library Python; no network.

Provides semantic memory and persistent storage for Claude, leveraging ChromaDB and sentence transformers for enhanced search and retrieval capabilities.

Enables agents to audit their own reasoning trajectories for loops, deadlocks, and logical contradictions, then perform non-destructive checkpoint rollbacks to a sound prior step. It combines step-level reward evaluation, MCTS search, symbolic proof checking, and consensus voting to return verified solutions over MCP stdio.

AI memory server with 53 tools. Knowledge Graph, semantic search, session tracking, multi-agent support.