mcp-serverman
a cli/mcp server tool for managing mcp server json config file with version control, profiles and multi-client support
Install / Use
claude mcp add benhaotang -- npx -y github:benhaotang/mcp-servermanIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
Customer SupportSupported Platforms
Tags
Skill content
View source on GitHubmcp-serverman: A MCP Server Configuration Manager
A command-line tool to manage Claude MCP servers configuration with version control and profiling. Now also has a companion mcp server to let llms config for you.
[!IMPORTANT]
I always recommend making a manual backup of the mcp configuration before making any changes. Although I tried to cover some error handling in the code, it is definitely not inclusive.
:floppy_disk:Installation
pip install mcp-serverman
or from GitHub for the latest debug version:
pip install git+https://github.com/benhaotang/mcp-serverman.git
Should be available on Windows, Linux(tested) and MacOS. If the path for a certain platform is wrong, open an issue.
:computer: Cli usage
After installation, you can use the mcp-serverman command directly in terminal:
# Display help message
mcp-serverman
# Initialize Client configuration(one time and must be done before using other commands, since 0.1.9)
mcp-serverman client init
# List servers
mcp-serverman list
mcp-serverman list --enabled
# Enable/disable/remove server/server version
mcp-serverman enable <server_name>
mcp-serverman disable <server_name>
mcp-serverman remove <server_name>
# Version control
mcp-serverman save <server_name> --comment <comment>
mcp-serverman change <server_name> --version <version>
# Preset/Profile management
mcp-serverman preset save <preset_name>
mcp-serverman preset load <preset_name>
mcp-serverman preset delete <preset_name>
# Multiple client support(since 0.1.9)
mcp-serverman client list
mcp-serverman client add <short_name> --name "Display Name" --path "/path/to/config.json" --key "mcpServers" [--default]
mcp-serverman client remove <short_name>
mcp-serverman client modify <short_name> --default
mcp-serverman client copy --from <short_name> --to <short_name> --merge
# Register companion mcp server to let Claude/LLM manage for you(since 0.2.1)
mcp-serverman companion [--client <client>]
For detailed usage instructions, see the manual.
:robot: Install as a mcp server
# Install companion (since 0.2.1)
mcp-serverman companion [--client <client>]
Example:
- What mcp servers do I have?
- Disable xxx, xxx server for me.
:wrench:Development
To install the package in development mode, clone the repository and run:
pip install -e .
:checkered_flag:Roadmap
- [x] Add support for other MCP-Clients, e.g. Cline and MCP-Bridge (since 0.1.9)
- [x] Update the code to be more modular and easier to maintain (since v0.2.0)
- [x] Added our own mcp-server to let Claude/LLM manage for you (since 0.2.1)
- [ ] Better error handling tests
- ~~Integration with other MCP server install tools, e.g. Smithery, or with predefined installation templates (should iron out safety issues first)~~ We have so many mcp marketplaces now, just use mcp-installer package and instruct the model to install it for you.
- ~~Maybe a Web UI via Flask?~~
License
MIT License (LICENSE)
Related Skills
Agent-Reach
84.4kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
headroom
73.4kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
ruflo
73.0k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
CowAgent
47.1kOpen-source super AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, multi-channel. Lightweight, extensible, one-line install.
