tool-creator

Create and upgrade executable Python tools decorated with @tool.

630|68|Updated Mar 13, 2026
One-click install
npx skills add https://github.com/AgentTeam-TaichuAI/ScienceClaw --skill tool-creator-agentteam-taichuai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tool-creator
Source: https://github.com/AgentTeam-TaichuAI/ScienceClaw/tree/main/ScienceClaw/backend/builtin_skills/tool-creator
Command: npx skills add https://github.com/AgentTeam-TaichuAI/ScienceClaw --skill tool-creator-agentteam-taichuai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Create and upgrade executable agent tools to extend capabilities and adapt to user needs.

Core Features & Use Cases

  • Enable users to implement new tools as Python functions decorated with @tool, and upgrade existing tools.
  • Provide a standardized workflow for testing in the sandbox and saving via propose_tool_save to the Tools directory.
  • Supports converting scripts into reusable tools and managing the tool lifecycle across development and production.

Quick Start

Create a new @tool function in Python, test it in the sandbox, then save it to the Tools directory.

Frequently Asked Questions about tool-creator

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

FAQPage Schema
How do I create a reusable Python tool for an agent?

To create a reusable Python tool, implement a new function decorated with @tool, test it in a sandbox, and save it to the Tools directory via the propose_tool_save workflow.

What are the requirements for agent tool creation?

Agent tool creation requires one tool per file, descriptive docstrings, and type hints to ensure proper execution and lifecycle management within the sandbox environment.

Can I convert existing Python scripts into agent tools?

Yes, you can convert existing scripts into reusable agent tools by wrapping the logic in a Python function decorated with @tool and following the approved save workflow.

How do I test and upgrade executable tools in a sandbox?

You test and upgrade executable tools by running the modified @tool Python functions within the sandbox environment, then saving the finalized versions to the Tools directory.

Does agent tool creation support managing the tool lifecycle across environments?

Yes, it supports managing the tool lifecycle across development and production by enforcing standardized testing in the sandbox and saving approved tools via propose_tool_save.

What is the best way to manage tool lifecycle in a sandbox environment?

The best way to manage tool lifecycle in a sandbox is to develop @tool Python functions, test them thoroughly, and use propose_tool_save to store them in the Tools directory.