tools

Invoke JSON Schema-defined tools across OpenAI, Anthropic, Gemini, and Ollama.

3|Updated Aug 17, 2025
One-click install
npx skills add https://github.com/juanre/llmring --skill tools
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tools
Source: https://github.com/juanre/llmring/tree/main/skills/tools
Command: npx skills add https://github.com/juanre/llmring --skill tools

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Integrating external functions or APIs with LLMs for agentic behavior is often provider-specific and complex. This Skill provides a unified way to define and execute tools, simplifying agent development and enabling powerful automation.

Core Features & Use Cases

  • Unified Tool API: Define functions using JSON Schema once and use them consistently across OpenAI, Anthropic, Google, and Ollama.
  • Automated Function Calling: Enable LLMs to intelligently decide when and how to call external tools based on user prompts, extending their capabilities.
  • Multi-turn Execution: Manage the full lifecycle of tool use, from the model's request to tool execution and result integration back into the conversation.
  • Use Case: Create an AI assistant that can check the weather, search the web, or book appointments by automatically calling the appropriate external APIs, all through natural language commands.

Quick Start

Use the tools skill to ask the 'tool-user' alias "What's the weather in NYC?" with a 'get_weather' tool defined.

Frequently Asked Questions about tools

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

FAQPage Schema
How do I enable LLMs to call external functions automatically?

Tool use lets LLMs intelligently invoke external functions based on user intent. Define tools using JSON Schema, then the model decides when and how to call them—enabling capabilities like weather lookup, web search, or API integration without manual routing logic.

Can I use the same tool definitions across OpenAI, Anthropic, and other LLM providers?

Yes. This Skill provides a unified JSON Schema–based tool API that works consistently across OpenAI, Anthropic, Google Gemini, and Ollama, eliminating provider-specific tool definition syntax and letting you switch providers without rewriting tool logic.

What's the full workflow for tool invocation in multi-turn conversations?

Tool use handles the complete lifecycle: the model requests a tool call, you execute the external function, return the result to the conversation, and the model incorporates it into its response. This orchestration repeats across turns, enabling stateful agent behavior.

Do I need to manually parse tool responses and feed them back to the model?

No. This Skill manages tool result messaging automatically—capturing tool outputs, formatting them for the model, and integrating responses back into the conversation flow without manual intermediary steps.

How do I control whether the model must call a tool or can choose to skip it?

Use the tool_choice parameter to enforce tool invocation, allow optional use, or disable tools entirely. This gives you fine-grained control over agentic behavior—from mandatory function calls to models that decide independently whether a tool is needed.