tmux-mcp
Active·★ 57·MIT·Updated 2026-07-11
tmux-mcp-rs is a Rust-based Model Context Protocol (MCP) server that enables AI agents to programmatically interact with and manage tmux sessions. It allows agents to create, split, and run commands within tmux panes while providing structured output and real-time user observation, making terminal automation more reliable and efficient.
#tmux#AI Agent#Workflow Automation#Terminal Management#Rust#Model Context Protocol (MCP)#Shell Automation#Remote Control
01
Features
01AI Agent Control: Allows AI assistants to create, manage, and interact with tmux sessions, windows, and panes.
02Structured Interaction: Provides structured inputs and outputs for tmux commands, reducing ambiguity for agents compared to plain-text skills.
03Command Execution & Tracking: Enables execution of commands within panes and tracking their results, including exit codes and attributable output.
04Pane Content Capture & Search: Can capture pane content and offers advanced buffer search capabilities for large contexts.
05Remote & Isolated Operation: Supports controlling tmux sessions over SSH and using isolated sockets for agent-specific environments.
02
Compatibility
tmux
Runtime
Verified via docs
Rust
Language
Verified via docs
Linux
Operating System
Verified via docs
macOS
Operating System
Verified via docs
SSH
Connection
Verified via docs
03
Quick start
1
$ cargo install tmux-mcp-rs
04
Use cases
↳Automating Interactive Terminals: AI agents can create, manage, and operate within their own tmux sessions, allowing users to observe or participate in real-time.
↳Reliable Shell Automation for AI: Agents can execute complex shell commands and workflows, benefiting from structured outputs and stable IDs for sessions/windows/panes.
↳Incremental Context Exploration: Using buffer search tools, agents can explore large text contexts (like logs or codebases) incrementally without loading everything into LLM context.
↳Agent Orchestration and Parallel Execution: Running parallel commands across multiple windows/panes with monitoring for complex tasks.
↳Hardened Agent Sandboxing: Enforcing security policies, command filtering, and restricting agent access to specific sockets, sessions, or panes.
05
Alternatives
codflow★ 377
The open-source, COD-first e-commerce + delivery platform for Algeria built agentic-ready.
showtime★ 273
A local video studio for your coding agent. Describe a video in one sentence: your agent directs, your machine renders. Claude Code, Codex, Cursor, Devin and more. No API keys, no uploads.
superCMO-skills★ 138
Open-source skills that empower any AI agent (Claude, Cursor, Hermes, etc.) to generate end-to-end marketing campaigns - UGC videos, ad videos, product photography and more.
dcc-mcp-core★ 70
Core runtime and shared tooling for DCC-MCP: safe, skill-first MCP integration across Maya, Blender, Houdini, Photoshop, and custom DCC hosts.
Comments
Log in to leave a comment
No comments yet. Be the first!