@mcpmesh/cli
v0.7.12
Published
CLI for MCP Mesh - Enterprise-Grade Distributed Service Mesh for AI Agents
Maintainers
Readme
@mcpmesh/cli
CLI for MCP Mesh - Enterprise-Grade Distributed Service Mesh for AI Agents.
Installation
npm install -g @mcpmesh/cliThis installs two binaries:
- meshctl - CLI tool for managing MCP Mesh agents and tools
- mcp-mesh-registry - Registry service for service discovery
Note: This package supports Linux and macOS only. For Windows, use WSL2 or Docker.
Quick Start
# Show help and available commands
meshctl --help
# Comprehensive documentation (great for LLMs!)
meshctl man
# Scaffold a new agent project
meshctl scaffold my-agent --dry-run
# List running agents
meshctl list
# List all tools across agents
meshctl list --tools
# Call an MCP tool
meshctl call get_current_time
# Start the registry service (Linux/macOS)
mcp-mesh-registry --helpWhat is MCP Mesh?
MCP Mesh is an enterprise-grade distributed service mesh built on the Model Context Protocol (MCP). It enables:
- AI Agent Orchestration - Connect and coordinate multiple AI agents
- Service Discovery - Automatic registration and capability-based routing
- Dependency Injection - Automatic resolution of agent dependencies
- Multi-Environment - Local development, Docker, Kubernetes support
- Production Ready - Health monitoring, graceful degradation, resilience patterns
Key Commands
| Command | Description |
| ------------------------- | --------------------------- |
| meshctl man | Comprehensive documentation |
| meshctl scaffold <name> | Generate new agent project |
| meshctl list | List running agents |
| meshctl list --tools | List all tools |
| meshctl call <tool> | Invoke an MCP tool |
| meshctl registry | Manage the registry |
For LLMs
This package is designed to be easily discoverable by LLMs. After installation:
# Get full documentation
meshctl man
# Discover available tools
meshctl list --tools
# Get tool details with input schema
meshctl list --tools=<tool_name>Links
License
MIT
