"filesystem" matching MCP servers:

A Model Context Protocol server that provides AI agents with secure access to local filesystem operations, enabling reading, writing, and managing files through a standardized interface.

A filesystem-based MCP server for AI coding agents to coordinate work across git worktrees by claiming files, checking for conflicts, and logging progress without affecting the repository's git history.

Enables local, event-driven messaging between multiple AI coding-agent sessions on the same machine using only Node.js and the filesystem, allowing them to hand off tasks, exchange replies, and broadcast announcements without external infrastructure.

Local inter-agent messaging for AI coding agents via filesystem relay.

Lets a frontier coding agent delegate research, cataloguing, and long-running computation to a local LLM with guarded filesystem, web, and Python execution tools, preserving the agent's context and tokens.

Bridges MCP (Claude Code) to ACP (Devin), enabling Claude Code to delegate autonomous tasks to Devin with permission and filesystem sandbox controls.

Enables primary AI orchestrators to delegate token-heavy coding tasks to economical models via OpenRouter, running sandboxed autonomous ReAct loops with filesystem operations and git audit tools while reducing token costs.

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

Policy-enforcing MCP proxy that blocks dangerous tool calls before they execute. Protects credentials, filesystem, shell, and databases across Claude Desktop, Cursor, Windsurf, and OpenClaw.

Enables durable, crash-safe control of long-running AI-agent workflows with idempotent operations and scope-aware reconciliation over MCP filesystem and execution tools.

Enables AI coding providers to run structured multi-agent workflows through MCP by owning task claims, action logs, handoffs, health gates, and specification metadata. It coordinates use cases, specs, implementation, review, and bounded repair while keeping document files on the filesystem.

Airlock provides named resource locks, atomic shared state, presence, events, and a task queue for AI agents to coordinate on a shared filesystem, with blocking waits and SQLite-backed persistence.

Enables AI coding providers to coordinate structured multi-agent workflows with documents, tracked tasks, handoffs, and health evidence. It owns task claims, action logs, and final health results while leaving specification files on the filesystem.

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

Enables MCP-capable agents to execute generated Python, Node.js, or shell code inside disposable, network-isolated Docker containers with strict resource, filesystem, privilege, and time limits, returning structured results and errors.

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.

Enables local AI agents to collaborate as Driver and Navigator for pair programming and technical peer review, exchanging proposals, diffs, and messages through a zero-cloud filesystem-backed MCP server.

Enables AI coding providers to run structured multi-agent workflows through MCP with tracked tasks, action logs, handoffs, health gates, and filesystem-backed specification documents.

Enhanced MCP server for full computer control: mouse, keyboard, screenshots, OCR, deep UI automation, semantic element actions, process management, filesystem watching, and accessibility-aware app launching.

Enables autonomous software-engineering tasks in a repository by coordinating filesystem, GitHub, database, and CI/CD tools through MCP, with a deterministic safety boundary for permissions, validation, timeouts, duplicate protection, and test verification.

Executes TypeScript code in a secure sandbox with IPC-bridged access to MCP tools like filesystem operations and network requests.

Enables AI agents to perform narrow, validated Fritzing engineering operations — searching parts, inspecting connectors, placing and moving components, wiring, validating, rendering, and saving projects — through a strictly sandboxed server that exposes no shell, arbitrary filesystem access, or direct Fritzing control, with lifecycle gating and a part trust model enforcing validation before render and save.

Provides a provider-agnostic MCP server that runs structured multi-agent workflows by owning task claims, action logs, handoffs, and server-side health gates over a shared filesystem and database. It enables coordinated Explorer, Consultant, Builder, and Reviewer roles with tracked specifications and audited completion evidence.

A universal interceptor that sits between any AI model's tool calls and shell execution, routing them through a supervisor daemon that inspects AST deltas, subprocess output, and hash-chained ledgers before work is allowed. It enables users to kill deleted assertions, trivial stubs, phantom test runs, and cyclic self-debate loops deterministically at the compiler, process, and filesystem boundaries rather than via prompt instructions.

Enables AI coding providers to run structured multi-agent workflows with tracked tasks, handoffs, document specifications, and server-owned health gating and evidence. It coordinates roles like explorer, consultant, builder, and reviewer while keeping documents on the filesystem and task state in a database.

A local stdio MCP server that lets Claude Desktop hand tasks to a Hermes Agent on the same machine and read the raw output and files back, job by job. Long work runs as background jobs that are polled by id, so Claude can plan, verify against the filesystem, and resume earlier sessions without blocking on tool timeouts.

Exposes an MCP workflow that lets AI coding providers claim and track tasks, log actions, coordinate lead/explorer/consultant/builder/reviewer handoffs, and manage specification documents on the filesystem. Completion is gated on server-owned health checks, so finished work requires passing verification evidence rather than a provider's own claim.