---
title: "MCP Tool | AgentRQ Glossary"
url: https://agentrq.com/glossary/mcp-tool
description: "An MCP tool is a callable function exposed by an MCP server that an AI agent can invoke to perform actions — such as sending a message, creating a task, or reading data from an external system."
---

[Back to Glossary](https://agentrq.com/glossary/)

# MCP Tool

An **MCP tool** is a callable function exposed by an [MCP server](https://agentrq.com/glossary/mcp-server) that an AI [agent](https://agentrq.com/glossary/agent) can invoke to perform actions in the world. Tools are the primary mechanism through which agents extend their capabilities beyond text generation — enabling them to interact with external systems, APIs, file systems, and services.

## Anatomy of an MCP Tool

Each MCP tool has:

- → **Name** — A unique identifier (e.g., `task`, `memory`)
- → **Description** — Natural language explanation of what the tool does (used by the LLM to decide when to call it)
- → **Input schema** — JSON Schema defining the parameters the tool accepts
- → **Response** — The data returned to the agent after the tool executes

## AgentRQ MCP Tools

AgentRQ exposes the following tools to [Claude Code](https://agentrq.com/glossary/claude-code) agents:

| Tool | Actions | Replaces |
| --- | --- | --- |
| `task` | create, get, updateStatus, reply, elicit, getAttachment | `createTask`, `getTask`, `updateTaskStatus`, `reply`, `elicit`, `getAttachment` |
| `memory` | load, save, delete | `loadMemory`, `saveMemory`, `deleteMemory` |
| `skills` | search, load, save, delete | `searchSkills`, `loadSkill`, `saveSkill`, `deleteSkill` |
| `workspace` | get, publishEvent | `getWorkspace`, `publishEvent` |
| `siteTools` | search, getDefinition, call | `listSiteTools`, `getSiteToolDefinition`, `callSiteTool` |
| `pano` | save, read, comment, attach, delete | — |
| `help` | — | — |

Every tool takes two inputs, an `action` and a `payload` with that action's fields:

text

```text
task {"action": "create", "payload": {"title": "Review the release notes", "body": "Check the notes before Friday.", "assignee": "human"}}
```

The `help` tool prints the exact payload: `help {}` lists the tools, `help {"action": "task"}` lists one tool's actions, and `help {"action": "task.create"}` shows one action's fields and an example. The old names in the Replaces column still answer for clients that are already installed.

## How the Agent Uses Tools

When a tool is available, the LLM can choose to call it by generating a structured tool call in its response. The [MCP client](https://agentrq.com/glossary/mcp-client) intercepts this, executes the tool call against the server, and returns the result to the model as a new context entry.

text

```text
1. Model generates: task {"action": "reply", "payload": {"taskId": "...", "text": "Done!"}}
2. MCP client calls AgentRQ's task tool
3. AgentRQ delivers the message to the human's task board
4. MCP client returns: { "success": true } to the model
5. Model continues reasoning with updated context
```

## Related Terms

- → [MCP](https://agentrq.com/glossary/mcp)
- → [MCP Server](https://agentrq.com/glossary/mcp-server)
- → [MCP Client](https://agentrq.com/glossary/mcp-client)
- → [Tool Use](https://agentrq.com/glossary/tool-use)
- → [Task](https://agentrq.com/glossary/task)