tools

Expose MCP server tools as discoverable primitives via tools/list and tools/call.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/theslashdojo/dojo --skill tools-theslashdojo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tools
Source: https://github.com/theslashdojo/dojo/tree/main/nodes/mcp/server/tools
Command: npx skills add https://github.com/theslashdojo/dojo --skill tools-theslashdojo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mcp, and includes scripts (resource) components.

What problem does it solve?

MCP servers need a standard, scalable way to expose AI-accessible functions. This Skill provides a robust pattern for defining, discovering, and invoking model-controlled tools within an MCP server.

Core Features & Use Cases

  • Automatic tool discovery via tools/list and safe invocation via tools/call for dynamic AI actions.
  • Cross-language support with Python decorators (@mcp.tool()) and TypeScript equivalents (server.tool()) that generate input schemas from type hints.
  • Use cases include API wrappers, data processing, computation tasks, and workflow automation on MCP servers.

Quick Start

Define a new tool by decorating a Python function with @mcp.tool() (or a TypeScript tool) and run the MCP server so the tool becomes available to LLMs.

Frequently Asked Questions about tools

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I expose Python functions as discoverable MCP server tools for AI models?

You expose Python functions as MCP tools by decorating them with @mcp.tool(). This generates input schemas from type hints, enabling automatic discovery via tools/list and safe invocation via tools/call for AI models.

What is the standard way to handle tool discovery and invocation on an MCP server?

MCP server tool discovery and invocation uses the tools/list and tools/call primitives. This pattern allows AI models to dynamically find available functions and execute them safely using validated input and output schemas.

Can I define MCP tools in TypeScript the same way as Python?

Yes, you can define MCP tools in TypeScript using server.tool() equivalents. Similar to Python decorators, TypeScript tooling generates input schemas from type definitions to support automatic discovery and invocation.

Do I need a configured MCP server context before defining tools?

Yes, defining and exposing tools requires a configured MCP server context, either Python FastMCP or TypeScript tooling. You must run the MCP server so your defined tools become available to LLMs.

What are the limitations when using MCP server tools for workflow automation?

MCP server tools require proper input and output schemas with validation and safeguards to function correctly. They are limited to environments with a configured MCP server context and do not work without established schema definitions.

What is the best way to make API wrappers discoverable by LLMs on an MCP server?

Making API wrappers discoverable by LLMs involves defining them as MCP server tools using Python or TypeScript decorators. This applies standard input schemas, allowing automatic discovery via tools/list and execution via tools/call.