agent-connector
@ken-jo
基本情報
agent-connector について
Package-first framework for building branded MCP packages that install across agent CLI hosts and expose local-first token telemetry.
概要
What is agent-connector?
A package-first framework for MCP developers. It allows you to build your own branded MCP package by depending on @ken-jo/agent-connector, which provides cross-agent installation, host coverage, doctor checks, packaging, and local-first token telemetry.
How to use agent-connector?
Install the package with npm install @ken-jo/agent-connector, then use the provided CLI to package your connector. For example, run npx @ken-jo/agent-connector package --connector ./src/connector.ts --format all --out ./dist-plugin to generate distribution files.
Key features of agent-connector
- Package-first development for branded MCP tools
- Cross-agent installation infrastructure
- Host coverage and doctor checks for connectivity
- Automated packaging into multiple formats
- Local-first token telemetry
Use cases of agent-connector
- Build and distribute a custom MCP server package under your own brand
- Package a connector for use across multiple agent hosts
- Diagnose and verify MCP connection health with built-in doctor checks
FAQ from agent-connector
Is agent-connector an MCP server?
No. It is developer tooling and client-side installer infrastructure, not a single hosted MCP server.
What does the package command do?
It takes a connector file and outputs packaged plugin files. The example uses --format all to generate all supported formats into a specified output directory.
Where can I find full documentation?
Visit the website at https://agent-connector.ai/ or the agent-readable summary at https://agent-connector.ai/llms.txt. For complete context, see https://agent-connector.ai/llms-full.txt.
「エージェントフレームワーク」の他のクライアント
GenAIScript
microsoftAutomatable GenAI Scripting
MCP Agent Graph (MAG)
keta1930MCP Agent Graph (MAG) is an agent development framework for rapidly building agent systems. This project is based on graphs, nodes, and MCP to quickly construct complex Agent systems.
Unified MCP Client Library
pietrozullomcp-use is the easiest way to interact with mcp servers with custom agents
Novu - Connect your AI agents & products to customers
NovuNovu is the open-source communication layer for AI agents. An agent built on Novu connects to MCP servers like Linear, GitHub, and Notion for its tools, then talks with your users on Slack, Microsoft Teams, WhatsApp, Tel
コメント