メニュー
Provides a specialized AI agent for code generation, debugging, refactoring, and code review. Use for: automating coding tasks, improving code quality, and accelerating software development workflows.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
A comprehensive collection of templates, design patterns, architectural sketches, and documentation for building state-of-the-art autonomous AI agents. Use for: rapidly prototyping and deploying agentic workflows, understanding core agentic behaviors, and learning best practices in AI agent development.
Enables an AI agent to conduct deep-web research, synthesize information, and generate comprehensive reports. Use for: automated information gathering, literature reviews, market analysis, and report generation.
Provides a foundational template for building AI agents using the ReAct pattern. Use for: quick prototyping of new agent ideas, understanding core agentic loops, and as a starting point for custom agent development.
Provides a framework for coordinating multiple specialized AI agents to achieve complex goals. Use for: managing workflows involving diverse agent capabilities, task decomposition, and collaborative problem-solving.
Provides a template for building Retrieval-Augmented Generation (RAG) agents. Use for: creating agents that can access and reason over external knowledge bases, semantic search, and contextual retrieval.
| name | coding-assistant |
| description | Provides a specialized AI agent for code generation, debugging, refactoring, and code review. Use for: automating coding tasks, improving code quality, and accelerating software development workflows. |
This skill provides a specialized AI agent designed to act as a Coding Assistant, capable of assisting with various software development tasks.
To utilize this coding assistant skill, follow these steps:
class CodingAssistantAgent:
def __init__(self, llm, code_executor=None, linter=None):
self.llm = llm
self.code_executor = code_executor # Optional: for running and testing code
self.linter = linter # Optional: for static code analysis
def assist_coding(self, task_description, code_context=None):
# Logic for understanding the task, generating/modifying code,
# and potentially using code_executor or linter for validation.
pass