tool-calling

Define LLM functions and manage tool calls for structured JSON outputs.

26|2|Updated Jun 8, 2026
One-click install
npx skills add https://github.com/xberg-io/plugins --skill tool-calling-xberg-io
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tool-calling
Source: https://github.com/xberg-io/plugins/tree/main/plugins/liter-llm/.cursor-plugin/skills/tool-calling
Command: npx skills add https://github.com/xberg-io/plugins --skill tool-calling-xberg-io

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill helps define functions and tools for an LLM to call, ensuring structured JSON outputs, and facilitates handling tool schemas and response formats.

Core Features & Use Cases

  • Function Definition: Define custom functions for LLMs to call, with parameters and descriptions.
  • Structured Output: Generate structured JSON responses from LLMs, useful for data processing and automation.
  • Use Case: Use this Skill to extract specific data points from text, like names and ages, and output them in a structured format for easier analysis.

Quick Start

Call the 'get_weather' function with the location 'Berlin' to get the weather information.

Frequently Asked Questions about tool-calling

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

FAQPage Schema
How do I get structured JSON output from an LLM instead of plain text?

To get structured JSON output from an LLM, you define custom functions for the model to call. This framework manages tool schemas and response formats to ensure structured data extraction from text responses.

What is the best way to define custom functions for LLM tool calls?

The best way to define custom functions for LLM tool calls is by specifying parameters and descriptions within a structured schema. This allows the LLM to understand the function definition and execute it accurately.

How do I extract specific data points like names and ages from text responses?

You extract specific data points like names and ages by defining an LLM function to parse the text. The framework outputs the extracted information in a structured JSON format for easier automated processing and analysis.

Do I need JSON parsing capabilities to handle structured LLM outputs?

Yes, you need JSON parsing and structured data handling capabilities to process the LLM outputs. The framework generates structured JSON responses that require parsing to integrate into automated data processing workflows.