Loading connector details…
Loading connector details…
Choose a unique username to continue using AgentHotspot
by nicozumarraga • Automation & Orchestration
A lightweight, configuration-driven framework for building and orchestrating hierarchical AI agents using the Model Context Protocol (MCP), ideal for composing capabilities and delegating tasks across agents.
Delegate complex, multi-step research or reasoning tasks to specialized subordinate agents via MCP capabilities.
Expose or aggregate external tools (search, web APIs, custom tool servers) and make them available to other agents in a hierarchy.
A configuration-driven way to define LLM-backed capabilities and run agents as MCP servers (including integration with Claude Desktop).
Light MCP Agents lets you define agents, their tools, and reasoning capabilities via simple JSON configs and connect them over MCP so higher-level agents can delegate work to specialized lower-level agents. It supports running agents in server or client mode, automatic tool and capability discovery, and tool aggregation across agent hierarchies. The project is designed to be composable and extensible, enabling recursive reasoning workflows, capability delegation, and easy sharing of agent configurations (including integration with Claude Desktop). Example configurations and multi-agent demos are included to help you get started quickly.
Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.