Human-in-the-Loop Task Manager for AI Agents
Back to Glossary

MCP Server

An MCP server is a service that exposes tools, resources, and prompts to AI agents via the Model Context Protocol. Servers are the capability provider side of the MCP architecture — they define what actions an agent can take and respond to tool call requests from MCP clients.

What an MCP Server Provides

MCP servers expose three types of capabilities:

AgentRQ as an MCP Server

AgentRQ is a hosted MCP server designed for human-in-the-loop agent workflows. It exposes tools that allow Claude Code agents to communicate with their human principals in real time:

Tool Actions Description
task create, get, updateStatus, reply, elicit, getAttachment Work with tasks and talk to the human in them
memory load, save, delete Notes that outlive a task
skills search, load, save, delete The workspace's skills
workspace get, publishEvent Workspace context and events
siteTools search, getDefinition, call Tools of websites shared from Chrome
pano save, read, comment, attach, delete Markdown docs shared across the account's workspaces
help — Each tool's actions and each action's payload

Every tool takes an action and a payload. The older tool names, such as createTask and getTask, still work for clients that are already installed.

MCP Server Transports

Transport Use Case
stdio Local servers running on the same machine
HTTP + SSE Remote servers accessible over the network

AgentRQ uses HTTP + SSE transport, which means it can be accessed from any machine where Claude Code runs.

Building Your Own MCP Server

The MCP specification is open. You can build custom MCP servers in any language to expose your own APIs, databases, or services to AI agents. The Anthropic MCP SDK is available for TypeScript, Python, and other languages.

Related Terms

Start Free