"Best open-source NPX or UVX tools for terminal deployment on Windows" matching MCP servers:

An agent-capable MCP server with terminal access, filesystem control, and persistent memory. It includes sandboxing, access control, activity logging, and a file-backed memory store for context across conversations.

Local-first memory layer for AI coding agents — Markdown as source of truth, MCP server + CLI, human-reviewed capture, team knowledge via git PRs

Read-only MCP server providing AI access to verifiable web, GitHub, and local sources, plus a managed fantasy entity catalog, with strong security and provenance tracking.

Enables persistent, project-scoped memory for OpenAI Codex Desktop and CLI on Windows, capturing prompts and responses through managed hooks, with federated recall across projects and optional local graph extraction.

Open Kioku is the local-first code intelligence and evidence layer for AI coding agents (Claude, Cursor, Codex, Zed, and other MCP clients). Instead of letting agents blindly crawl files, guess blast radius, or read the entire repository repeatedly, Open Kioku gives them a precise, pre-indexed code index to build evidence-backed plans before they edit, and verify boundary limits after they edit.

AI-native ontology engineering MCP server for OWL/RDF/SPARQL. Validate, query, diff, lint, version, and govern knowledge graphs via Oxigraph triple store.

Enables AI hosts to preserve and resume task and project state across conversations, process restarts, and worker or model replacement by storing structured records, snapshots, receipts, and bounded context views through a local stdio interface.

Stores what AI agents learn — thoughts, decisions, session events and repo facts — in PostgreSQL with pgvector and serves it back through versioned MCP tools with hybrid vector-plus-full-text retrieval and budgeted context packs. Enables multiple agents to share one scoped, role-isolated long-term memory, with deliberate promotion of distilled candidates into durable memories.

Self-hosted OAM memory with cited recall over your existing deployment. Local stdio adapter with six tools for recall, retaining interactions and documents, operation readiness, stored Profile reads and exact memory removal. Run with npx -y @rocke-dong/oam-mcp; requires Node.js 22 and your OAM user-scoped API key. Submitted by the OAM maintainer with AI assistance.

Exposes a local SQLite-backed long-term memory to AI agents as MCP tools, letting them add conversations and documents, record dated facts with provenance, search and assemble context, and edit, inspect, or forget what was stored. It runs with your own embedding and LLM keys, so memories stay readable, exportable, and portable as plain markdown.

Gives Claude Code a persistent, fully local long-term memory by automatically capturing session activity, indexing past conversations, projects, scheduled tasks and Obsidian notes, and recalling relevant context through hybrid keyword-plus-semantic search with progressive disclosure. It also writes session handoffs that are injected into later sessions, so work resumes with prior decisions and next steps already in context.

MCP server that builds a deterministic, source-traceable knowledge index of any codebase, enabling glossary lookup, code graphs, and exact-token search with every fact linked to its source file and line.

Gives Claude Code persistent, local-first project memory: it records sessions, compresses them into searchable observations and summaries, and injects a compact project brief plus prompt-time recall at session start. It also exposes tools for recalling observations and timelines, managing durable tasks and notes, and reading any project's context, all backed by a local SQLite file with compression run on a local LLM.

A self-hosted MCP knowledge server that gives agents a filesystem-shaped store of rules, context, skills, and granted connections — opened in full for standing rules and as an index otherwise, then listed, searched, and read on demand. Agents can only propose diffs and new entries (reviewable and correctable by the owner) and receive credentials for exactly the connections they're bound to.

Open-source MCP server for mem0 that enables memory storage and retrieval using local LLMs, self-hosted, and Docker-free.

Provides MSP support tools (ticket search, draft response, KB search, update) with a deterministic security guardrail that refuses to draft responses for security tickets based on content scanning, even if mislabeled.

Free Industry 4.0 / IIoT tools, no API key or account: lint MQTT Sparkplug B topics, check a Unified Namespace (UNS) for naming and hierarchy problems, and run an industrial AI-readiness diagnostic for manufacturing digital transformation and smart-factory projects. Works in Claude Code, Codex and any MCP client.

Provides customer lookup, transaction history, knowledge-base search, and ticket management tools to any MCP client.

Enables semantic search across Apple Mail, Messages, Calendar, and Contacts on macOS using natural language queries. All processing happens locally with privacy-first vector indexing for fast similarity search.

Personal knowledge graph with 16 MCP tools. Auto-links, deduplicates, tracks themes, synthesizes insights.

Enables multiple MCP agents to read and write a single user-controlled, local-first fact layer under one user identity with explicit scopes, preserving sources, versions, permissions, and audit history. Supports structured memories such as preferences, facts, task state, and decisions, plus structured handoff capsules so one agent can resume another's work by task ID.

Gives AI assistants persistent, shared memory using the Zettelkasten method, so they retain corrections, preferences, and decisions across sessions and across tools via local, hybrid search.

A self-hosted second brain MCP server that enables capturing thoughts with deduplication and semantic search using local embeddings and PostgreSQL with pgvector, all running on your own hardware.

Enables searching, reading, and appending change notes in an Obsidian vault through local MCP tools.

Enables Claude to access and manage a persistent, human-readable knowledge graph of neurons, with semantic search, memory consolidation, and local ownership.

Turns a codebase into a persistent knowledge graph so AI coding agents can answer structural questions about functions, call chains, routes, and cross-service links through graph queries instead of reading files one by one.

Creates and manages personality profiles using a 12-facet model based on 33 academic frameworks, stores them as OpenClaw-compatible SOUL.md/IDENTITY.md, and supports progressive profiling with confidence tracking.

Enables an MCP-compatible host to connect to a local event-planning work ledger, where users can create and revise missions, draft and save artifacts, manage tasks, decisions, outcomes, proposed memory, next actions, and handoff requests through twelve bounded local tools. It supplies the connection and role instructions while relying on the host's reasoning model and making no provider calls.

MCP server for Open WebUI Knowledge Bases – search and access your knowledge bases from Cursor, Claude Desktop, and other MCP clients.

Persists Claude Code sessions to Notion as team memory, enabling queryable recall of decisions, work state, and project architecture.

An MCP server for storing and retrieving thoughts, notes, and ideas with semantic vector search, enabling integration with Claude Desktop and other MCP clients.

Provides a privacy-first adaptive review engine with local SQLite storage, enabling item creation, due review retrieval, grading, and statistics via MCP tools.

Extends VS Code Copilot with MCP tools for persistent semantic memory, code indexing and RAG search, sandboxed Python/shell execution, web fetch/search, dynamic tool management via a web UI, and structured multi-step reasoning.

Enables natural language interaction with Obsidian vaults through an MCP server, providing hybrid search, file management, and AI-powered analysis.

A persistent, cross-session knowledge base for AI agents that indexes session history into a searchable SQLite database with full-text search, enabling recall of past sessions, stored knowledge, and summaries.

Provides a scalable knowledge graph implementation for Model Context Protocol using Elasticsearch, enabling AI models to store and query information with advanced search capabilities, memory-like behavior, and multi-zone architecture.

On Board is a local MCP server that gives multiple AI agents and IDEs a shared project memory, ticket queue, and handoff history, so agents can seamlessly continue each other's work. It supports agent-to-agent wake events, enabling autonomous workflows like reject-fix-resubmit cycles without human relay.

A MongoDB Atlas-based Notion-style knowledge base management MCP server that supports user workspaces, project management, page CRUD with tree structure, and text search.

Enables Codex projects to read Git-backed kickoff, acceptance, handoff, progress, and next-step context, and to append validated, audited progress events through scoped MCP tools. This supports visible agent progress, resumable sessions, and safer handoffs.

A collection of MCP servers for Apple macOS apps (Mail, Contacts, Notes, Memory, Messages, Calendar, Reminders) enabling AI assistants to read, search, create, and update data via JXA, SQLite, and EventKit.