·
@mcpc-tech
About ·
Build agentic-MCP servers by composing existing MCP tools.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"mcpc": {
"command": "npx",
"args": [
"jsr",
"add",
"@mcpc/core"
]
}
}
}Tools
No tools detected
Fetch the live tool list by running this server in a temporary sandbox using the button above.
Overview
What is MCPC?
MCPC is an SDK for building agentic MCP (Model Context Protocol) Servers by composing existing MCP tools. It is designed for developers who want to create portable, collaborative agents that work across any MCP client.
How to use MCPC?
Install the package via npm, pnpm, or Deno, or run directly with the CLI using npx. Use the mcpc() function to define dependencies, write an agent description, and create a server. You can also use the website at mcpc.tech to generate agents interactively or add the MCPC builder agent to your MCP client for AI-assisted setup.
Key features of MCPC
- Portability and agent interoperability across all MCP clients
- Simple composition and fine-tuning of existing tools
- Built-in MCP logging and OpenTelemetry tracing
- Skills support following the Agent Skills specification
- Flexible execution modes for different scenarios
Use cases of MCPC
- Create powerful agentic MCP tools by describing your vision and referencing community tools
- Fine-tune existing tools by modifying descriptions, parameters, or wrapping results
- Build multi-agent systems by defining each agent as an MCP tool
- Create custom coding assistants (e.g., a Codex/Claude Code fork)
- Organize domain knowledge using Agent Skills for on-demand loading
FAQ from MCPC
What is MCPC?
MCPC is an SDK for building agentic MCP Servers by composing existing MCP tools.
How do I install MCPC?
Install via npm: npm install @mcpc-tech/core, or via jsr: npx jsr add @mcpc/core. You can also run directly with npx -y @mcpc-tech/cli.
What execution modes does MCPC support?
MCPC supports five modes: agentic (interactive), ai_sampling (autonomous), ai_acp (coding agents), code_execution (sandboxed JavaScript), and code_execution_sampling (sandbox with sampling).
Can I use MCPC without writing code?
Yes, you can use the website at mcpc.tech to browse servers and generate agents, or add the MCPC builder agent to your MCP client for AI-assisted discovery and building.
Does MCPC support tracing?
Yes, MCPC has built-in MCP logging and OpenTelemetry tracing support.
Frequently asked questions
What is MCPC?
MCPC is an SDK for building agentic MCP Servers by composing existing MCP tools.
How do I install MCPC?
Install via npm: `npm install @mcpc-tech/core`, or via jsr: `npx jsr add @mcpc/core`. You can also run directly with `npx -y @mcpc-tech/cli`.
What execution modes does MCPC support?
MCPC supports five modes: agentic (interactive), ai_sampling (autonomous), ai_acp (coding agents), code_execution (sandboxed JavaScript), and code_execution_sampling (sandbox with sampling).
Can I use MCPC without writing code?
Yes, you can use the website at mcpc.tech to browse servers and generate agents, or add the MCPC builder agent to your MCP client for AI-assisted discovery and building.
Does MCPC support tracing?
Yes, MCPC has built-in MCP logging and OpenTelemetry tracing support.
Basic information
More Other MCP servers
MCP Go 🚀
mark3labsA Go implementation of the Model Context Protocol (MCP), enabling seamless integration between LLM applications and external data sources and tools.

Figs, Relationship Expert
Figs O'Sullivan, LMFTCouples therapist Figs O'Sullivan (LMFT, 17 years, endorsed by Sue Johnson, creator of EFT) inside your AI. Ask relationship questions answered from his clinical writing, name the pattern in a fight, get a three-step rep
MaxKB
1Panel-dev🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。

Glasswarp
GlasswarpSee and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.
FastMCP v2 🚀
jlowin🚀 The fast, Pythonic way to build MCP servers and clients.
Comments