"Local MCP servers for self-hosted AI assistants" matching MCP servers:

Enables AI agent hosts to exchange durable, local sender/receiver messages through a SQLite mailbox exposed as MCP tools (connect, send, receive, reply, ack, status, disconnect), so a coordinator can dispatch commands and an executor can return results without model polling while idle. A configured host wake adapter is still required to turn an incoming event into an Agent turn.

Enables agents to reflect on their own thinking by returning a metathought question based on a deterministic selection from curated cognitive lenses.

A self-hosted MCP server that lets Claude (or any MCP client) drive Autodesk Fusion 360 on your own machine.

Provides a local MCP bridge for AI assistants to control a Mac by observing screen state and performing actions like mouse movement, clicking, typing, and opening URLs.

Enables MCP servers to enforce their own Proof of Delegation by refusing tool calls when authorization is revoked, suspended, or expired, with optional self-check and verification tools.

A local-first MCP memory server that gives AI coding agents persistent memory with hybrid semantic and keyword retrieval, working fully offline.

Enables AI agents to learn from their work by recording tasks, extracting patterns, detecting mistakes, and proactively surfacing insights, all using the agent's own model through a cooperative intelligence pattern.

Self Healing Infrastructure - MCP server providing AI-powered tools and automation by MEOK AI Labs

Lets a coding agent hand a plain-language app idea to a hosted AI software team that shapes, designs, builds, tests and deploys it, returning a one-time link into the full build workspace. Also lists your apps with their stage, pending approvals and live addresses after a standard OAuth sign-in.

Captures user corrections and improvements during interactions, logs them, finds recurring patterns, and prescribes preventive fixes like CLAUDE.md rules, skills, scripts, or MCP tools to avoid repeating mistakes.

A self-updating MCP system that analyzes design and code inputs with an LLM, refines them through conversation, and crystallizes approved patterns into a local skill library for MCP agents.

A generic OpenAI-compatible subagent MCP server that lets Claude Code delegate coding grunt-work to a local model via a tool-calling agent loop, with configurable concurrency and backend.

A universal MCP server that enables any LLM or AI agent to access expert skills from your local filesystem.

Enables Codex to delegate coding tasks to a locally hosted Qwen model through an MCP tool, which can read, write, patch, test, and inspect git diffs inside an allowed workspace.

Enables autonomous learning from interactions through pattern recognition and machine learning techniques. Continuously improves performance by analyzing tool usage, providing predictive suggestions, and sharing knowledge across MCP servers.

An AI-centric MCP server that enables automated Xilinx Vivado workflows, including project management, synthesis, implementation, and timing analysis. It allows AI agents to drive hardware design processes while integrating directly with the official Vivado GUI for visual context.

Enables any MCP client to delegate a scoped task to a local Ollama model that runs its own tool-calling loop for file operations and web search, with an allow-listed folder security model.

Enables multiple real Codex, Claude Code, and Z Code sessions to coordinate development through a shared role-governed task board with scope and worktree locks, an evidence-based quality workflow, and a web console, backed by PostgreSQL over MCP.

Provides Claude with persistent memory and learning capabilities through 10 automatic agents that capture decisions, errors, solutions, and patterns across conversations. Features an anti-compaction system to prevent context loss and enables infinite conversation continuity.

Identity infrastructure for AI agents. Gives agents an evolving persona, session continuity, and self-correcting retrieval so they stop being strangers. Local-first, model-agnostic.

Exposes eight MCP tools over streamable HTTP that let clients ingest alerts, list and diagnose incidents, analyze Terraform plan risk, propose remediations, and dry-run or execute policy-validated, approval-gated changes with automatic rollback.

Self-hosted AI orchestrator that monitors and manages homelab services, exposing them as MCP tools for LLMs to drive.

Enables compatible AI apps to observe and control a locally connected iPhone through stdio MCP tools for screenshots, taps, swipes, long presses, text entry, key and button presses, app listing, and app launching, all reusing the console's device session.

A local MCP server that lets Hermes supervise Claude Code, delegating focused coding, research, or review tasks to the Claude Code CLI and managing worker sessions, background jobs, cancellations, and read-only reviews.

A Model Context Protocol (MCP) server that supports STDIO, SSE and Streamable HTTP protocols for AI model interactions.

Enables local llama.cpp inference and dotbot agent workflows through MCP, providing one-shot chat completions and agentic tool loops.

Enables Claude Code on Windows to read the foreground window's UI Automation tree as structured text and then act on it with real, human-like SendInput mouse and keyboard events. It drives desktop apps through a perceive-decide-act loop without screenshots, a vision model, or any locally hosted LLM.

Self-custody AI agent wallets with passkeys, gasless transactions & programmable permissions (ERC-4337)

Multi-agent swarm orchestrator for AI workflows. Exposes blackboard read/write, agent dispatch, permission gating (AuthGuardian), token budget enforcement (FederatedBudget), and audit log tools over MCP. Supports 14 AI framework adapters including LangChain, AutoGen, CrewAI, Codex, and A2A — mix any frameworks in one swarm with race-condition-safe shared state.(Traffic light for agents)

Transforms HTML into Figma designs using only the user’s design system, ensuring correct component usage or safe fallback to primitives, while continuously learning from each build to improve mapping accuracy and consistency.