Can I use multiple LLMs with these agent patterns?

Yes, the skill includes a multi-model architecture pattern that allows you to use fast models for planning and more powerful models for complex reasoning or code generation.

What kinds of tools are covered in the schemas?

The skill defines standardized schemas for file operations (read, write, edit), code understanding (search, definitions), and terminal interactions.

How does this skill handle security and safety?

It provides patterns for granular permission levels (Auto, Ask Once, Ask Each, Never) and defines sandboxed execution environments to prevent unauthorized system access.

What is the core architecture of an autonomous agent?

It follows a Think-Decide-Act-Observe loop where the agent reasons about a task, plans a specific tool call, executes that action, and then observes the result to determine the next step.

Autonomous Agent Patterns

Name: Autonomous Agent Patterns
Author: claudiodearaujo

byclaudiodearaujo

•

데이터 과학 및 ML

Implement robust design patterns for building autonomous AI coding agents with integrated tool calling and safety permissions.

This skill provides a comprehensive framework for developing autonomous coding agents, drawing inspiration from industry leaders like Cline and OpenAI Codex. It covers the fundamental 'Think-Decide-Act-Observe' agent loop, multi-model selection strategies, and standardized tool schemas for file operations and terminal execution. Crucially, it includes sophisticated permission systems and sandboxing patterns to ensure AI agents operate safely and predictably within a development environment, making it essential for anyone building custom agentic workflows.

주요 기능

01Robust tool schemas for file manipulation and shell command execution

02Standardized Think-Decide-Act-Observe agent loop architecture

03Sandboxed execution patterns for secure command handling

04Multi-model orchestration for optimized planning and execution

051 GitHub stars

06Granular permission levels and human-in-the-loop approval workflows

사용 사례

01Designing secure tool APIs for LLM-based autonomous workflows

02Implementing safety gates and human approval steps for automated refactoring tasks

03Building a custom AI coding assistant tailored to a specific codebase

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add claudiodearaujo/sistema-de-narra-o-de-livro autonomous-agent-patterns

For use in Claude.ai and ChatGPT

Download Skill