"Creating files with a public endpoint" matching MCP servers:

Enables ChatGPT web to read local files, search code, run shell commands, and drive a local Codex CLI via an official MCP connector, with human approval for risky commands and full audit logging.

A minimal MCP server that exposes a run_command tool over stdio, enabling command execution within a Docker container.

Memory bank with Server as SSH support for central knowledge base

A local MCP bridge that lets ChatGPT chat mode operate the user's own computer, running project, file, and shell tasks through a local executor and returning progress and results to the conversation. It also coordinates Codex sessions (persistent or one-off) and other agent CLIs as sub-agents.

Enables bounded agent authority over a macOS workstation via 63 MCP tools with policy, approval, and audit.

A demo MCP server that enables users to interact with various tools through natural language, including note management, weather queries, calculations, file operations, and shell commands. Uses Ollama for intent detection and structured command parsing from free text input.

Enables remote server control via SSH through Claude, supporting command execution, file transfers, and multi-server management with both standard and streamable HTTP protocols.

Provides secure filesystem access for AI models through the Model Context Protocol with strict path validation, file operations, directory management, and system command execution within predefined directories.

Enables AI-assisted development tools like Cursor, Cline, and Windsurf to request user confirmation before acting, running shell commands and displaying their output inside a Qt feedback window that returns user notes and logs to the model. This human-in-the-loop loop lets the assistant consolidate many speculative tool calls into a single feedback-aware request, saving API usage and improving accuracy.

Enables ChatGPT Web conversations to operate a Windows computer through MCP with local tools, permissions, task state, and audit evidence.

Enables agents to spawn, monitor, and control CLI coding agents running in isolated tmux sessions, with verified input injection, redacted pane capture, fleet and per-session state classification, lineage tracking, blocking exec with exit codes, cross-agent messaging, waste analytics, and log/ledger reading. Ownership checks, policy controls, and audit events keep every control action safe by default.

A self-hosted MCP server that enables AI coding agents to read, edit, search, and run code in local projects with human review loops and policy controls.

Enables AI assistants to access files and terminal of a local computer via a public HTTPS endpoint, secured with GitHub OAuth.

A minimal MCP server for UW Hyak (klone) that exposes tools to run shell commands and write files remotely, along with documentation resources for klone-specific utilities.

A production-ready MCP server that enables AI assistants to execute shell commands, manage files, monitor system resources, and automate complex workflows with advanced features like stock tracking and web automation.

Enforces mandatory human approval before LLM agents can edit files, create files, or execute shell commands, using session tokens and audit logging.

Enables management of visible, interactive terminal sessions across platforms (macOS, Windows, Linux, WSL). Supports creating, executing commands, capturing output, and managing multiple terminal windows simultaneously.

Enables an AI agent to run shell commands and move files inside a WSL distro on the same Windows machine via wsl.exe, with no sshd, key, or port required. Each pre-registered distro-and-user entry enforces guard policies limiting dangerous commands, Windows interop, write targets, and accessible paths.

Enables MCP clients to securely work with local files, directories, searches, terminal sessions, and processes on paired computers via outbound connections, without exposing inbound ports.

Enables Claude to work with iTerm2 terminal sessions by staging commands for user approval, executing them after approval, creating new tabs, and returning only new output.