"Information on agent swarm architecture" matching MCP servers:

Connects ChatGPT to local files, terminals, and the desktop through MCP with Chrome integration, letting users read and edit project files, run commands, and follow tool results as they arrive. It also adds durable multi-agent workers, Goal/Loop task tracking, mid-task corrections, and Compact & Resume for carrying session and worker history into a fresh chat.

Enables ChatGPT to access approved local Windows folders, run commands, control the desktop, and manage session history through a permission-bounded MCP server.

A read-only remote MCP server that searches and reads files from a personal Microsoft OneDrive folder via Cloudflare Workers, enabling live access without a local workstation or tunnel.

Enables ChatGPT's web interface to operate a real Mac by executing shell commands and reading or writing files over a reverse SSH tunnel, without installing a client or consuming API quota.

Enables Cursor's AI agent to perform supercharged local file search on Windows by connecting to Agent Ransack via MCP, supporting content and filename searches, regex, and deep folder scans.

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.

MCP server that lets a private ChatGPT app remotely control a paired Windows PC, including files, PowerShell, programs, screenshots, mouse/keyboard, clipboard, and Chrome.

Enables coding agents to send files and notes into approved Zas channels and read items back, with content encrypted before leaving the machine and scoped, revocable agent access.

An MCP server exposing read, write, edit, patch, glob, grep, shell, webfetch, and websearch tools with OpenCode-compatible names and parameters over Streamable HTTP or stdio. It lets an agent work in a root-confined workspace on another machine, with optional bearer-token authentication and guarded private-network and out-of-root access.

Provides tools to monitor host system health including CPU load, disk usage, and network status while enabling file system management tasks like searching and moving files. It includes built-in safety guards to prevent operations on critical system directories.

Enables MCP clients to drive developer workflows through APIs and CLIs, with explicit workspace or trusted-workstation launch modes, bounded command previews, multi-repository rules discovery, and an on-demand skills/recipes entry point. It lets agents review changes, run commands, and load only the smallest relevant rule or skill while reporting evidence and unverified steps.

Enables AI assistants to analyze KTuner datalogs, check logged runs, and propose or write safe ECU table edits into a new .kcl tune file for review and flashing in KTuner. It helps identify knock, fuel, and ignition timing changes without flashing the car or altering the original tune file.

A read-only MCP server that lets local AI agents inspect open Mac windows, browser tabs, PDFs, clipboard contents, Finder selections, and Messages history so they can answer questions about what the user is currently viewing.

Enables agents to read and write Markdown documents, HTML pages, data files, PDFs, screenshots, and videos through a standard remote MCP server with OAuth, per-agent revocable project permissions, version history, and stable artifact links.

Provides a local markdown-based memory system for coding agents through MCP, enabling search, add, register, inventory, sync, and ingest operations across user and project memory. It gives any agent tool a durable, provider-agnostic shared memory stored in folders you own.

Enables AI agents to maintain project continuity through a file-based state hub with tasks, phases, and handoff snapshots. Provides MCP tools for reading and updating project state, with gatekeeping enforced via real-state evaluation and per-tool authorization.

Enables AI agents to automate a Linux desktop natively on Wayland, with 68 tools for screen capture, input simulation, browser automation, OCR, accessibility, media, audio, clipboard, notifications, filesystem, system, memory, and network operations through the MCP protocol.

Turns web-based AIs into full-powered agents with local PC access, 33 tools across 8 categories, and parallel swarm orchestration with real-time validation.

Archive Agent is an open-source semantic file tracker with OCR + AI search (RAG) and MCP capability.

Connects ChatGPT and Claude to your own machine so they can read and edit files, search paths and text, run shell, Git, build and test commands, and keep cross-conversation memory and reusable skills. It exposes these capabilities as 14 Model Context Protocol tools over a remote Streamable HTTP endpoint with API key or OAuth authentication, plus an operator dashboard for monitoring and configuration.

An MCP server that delegates filesystem operations to a specialized React agent, reducing context usage and improving accuracy for AI applications like Claude Code.

MCP server that enables AI clients to orchestrate multi-node LAN agents, with tools for remote command execution, file transfer, task delegation, and cross-node collaboration.

Enables Notion AI to read and modify files in a single local PC folder and, in full mode, run commands and Codex, Claude Code, or Gemini CLI agents, with user confirmation for each action.

MCP server for the mimOE AI Agent, enabling on-device tools like device discovery, network insight, and persistent note storage via natural language.

A local MCP filesystem wrapper that previews file mutations, requires user approval, and records verifiable undo information for supported clients like Claude Desktop, OpenCode, and Codex.

Local automation bridge that is both an MCP server and client: 234 tools across 44 namespaces for shell, filesystem, git, desktop control, browser (CDP), Android (ADB), OCR/ASR and memory, plus the ability to connect external MCP servers. Sandboxed execution with token auth, fail-closed policy and audit log.

An MCP server that compiles any text into a verifiable, graph-based knowledge base using deterministic chunking and parallel extraction of epistemology primitives.

Local daemon that bridges AI assistants to the local filesystem with read-only commands and optional write mode, enforcing strict security filters to prevent credential leakage.

Enables local file exchange between users and CLI agents via a web UI and MCP server, allowing agents to read/uploads and deliver artifacts without copy-paste.

Enables users to chat with an LLM agent that creates and reads Word, Excel, PDF, and JSON files, and accesses databases (letters, documents, LIMS) according to user permissions.