uvwtDeveloper tools
agentdock
Secure MCP runtime for AI agents to operate local machines, servers, and containers with multi-device orchestration.
- Backend
- DevOps
- Full stack
- Automation
- Productivity
- Desktop
- CLI
- Windows
- macOS
- Linux
- Browser
- Self-hostable
- Runs locally
- Docker supported
- Popularity
- 1.2k Stars
- GitHub stars
- Recent activity
- 10/2/2026
- Updated in the last 30 days
- License
- APACHE-2.0
- Permissive
Why it matters
We look beyond stars: what problem it solves, whether it creates real utility, and what makes its approach worth noticing.
Problem
Addresses the difficulty of letting AI chat clients safely and directly manipulate multiple local and remote physical machines or servers for real-environment tasks.
Practical value
Exposes operation surfaces of multiple machines to chat clients via the MCP protocol, supporting files, Shell, Git, and multi-instance coordination with high practical utility.
Innovation / differentiation
Employs an independent tool runtime architecture, abstracting multi-device management into standard Model Context Protocol instances with clear boundary separation.
Leverage potential
Connects general conversation models with heterogeneous real environments, enabling cross-device complex configuration and deployment workflows from a single chat window.
Why now
Arrives during the rapid expansion of AI agents and the MCP protocol ecosystem, aligning with the demand to securely connect external tool capabilities to real machines.
Community activity
In the last 90 days there were 36 new issues and 145 pull requests; the bounded issue/PR samples include 29 issue authors and 18 PR contributors, with 23 releases.
Maintainer responsiveness
The 36-issue window sample had a 83% close rate, and the 100-pull-request sample had a 88% merge rate. Maintainer-response observations covered 100% of that issue sample, with a 64% response rate and median first response of 12.2 hours.
Key highlights
- Unified and secure file, command, Git, Skill, MCP, and browser automation execution
- Multi-device instance coordination to finish cross-device workflows in a single conversation
- Same tool model across macOS, Linux, Windows, and containers
Quick start
How it is installed, how hard it is, and where to start.
Where it runs
Runs locally
Difficulty
Easy — follow the steps
- 01Visit the latest releases page to download the installer for your operating system or use the automated script
- 02Complete the installation following the instructions for your platform (.exe for Windows, .dmg for macOS, install script for Linux, or Docker deployment)
- 03Choose a connection option (local only, temporary public address, or fixed domain), and get the MCP URL and authentication credentials
- 04Configure AgentDock into the MCP settings of your AI client (such as ChatGPT/Claude)
Best for
- Teams that want data on their own servers
- Developers who want to try it on their machine
- People who prefer Docker deploys
More about it
AgentDock is an independent tool runtime for AI agents.
It provides unified, secure, and controlled file, command, Git, Skill, MCP, browser automation, and task execution across local computers, remote servers, and containers. Connect multiple AgentDock instances to coordinate work across devices and finish multi-machine workflows in a single conversation.
AgentDock does not provide a chat interface or perform model inference. It focuses on one responsibility:
Let AI agents operate real environments within explicit permission boundaries and return structured, traceable, and verifiable results.
Sources
Each field shows its status and source — expand to review.
12 · Expand
Sources
Each field shows its status and source — expand to review.
capability tags
Verifiedautomation, productivity
Source: admin_cms · cms editor · 8/14/2026
Latest release
Verifiedv0.9.1
Source: GitHub API · latest_release=v0.9.1 · 10/2/2026
License
VerifiedApache-2.0
Source: GitHub API · license.spdx_id=Apache-2.0 · 10/2/2026
needs api key
VerifiedNo
Source: admin_cms · cms editor · 8/14/2026
One-liner
Verified{"en":"Secure MCP runtime for AI agents to operate local machines, servers, and containers with multi-device orchestration.","zh":"为 AI 助手提供安全、受控的访问权限，以操作您运行的每台机器。"}
Source: admin_cms · cms editor · 8/14/2026
Platforms
Verifiedwindows, macos, linux, browser
Source: admin_cms · cms editor · 8/14/2026
Category hint
VerifiedMCP Runtime
Source: description · The description calls it 'Secure MCP runtime for AI agents' and the README consistently refers to MCP (Model Context Protocol) as the tool exposure mechanism. · 8/14/2026
product forms
Verifieddesktop, cli
Source: admin_cms · cms editor · 8/14/2026
role tags
Verifiedbackend, devops, fullstack
Source: admin_cms · cms editor · 8/14/2026
supports docker
VerifiedYes
Source: Repository file · dockerfile=true; compose=true · 10/2/2026
supports local
VerifiedYes
Source: admin_cms · cms editor · 8/14/2026
supports self host
VerifiedYes
Source: admin_cms · cms editor · 8/14/2026
Related projects
Other verified projects matched by category, capabilities, and intended roles.
WSL
Windows Subsystem for Linux allows you to run Linux command-line tools and applications directly on Windows without the overhead of a traditional virtual machine.
opencodex
A universal local proxy that enables the use of any LLM, such as Claude, Gemini, or DeepSeek, within Codex and Claude Code, while managing ChatGPT account pools.
cli-printing-press
Automatically generates agent-optimized Go CLI tools and MCP servers by reading API documentation or sniffing web traffic.
deja-vu
A local memory layer for coding agents that indexes and retrieves past session decisions and code edits.
omni
Reduces AI Agent input token consumption by replacing redundant file reads and command outputs with compact, retrievable handles.
Flare
A graph-first IDE for agentic coding featuring a live codebase map, integrated terminal, and MCP task collaboration.