Skip to main content

Ce Agent Native Architecture

Skill Verified Active

Build applications where agents are first-class citizens. Use this skill when designing autonomous agents, creating MCP tools, implementing self-modifying systems, or building apps where features are outcomes achieved by agents operating in a loop.

Purpose

To guide developers in building sophisticated agent-native applications by establishing core architectural principles and practical patterns.

Features

  • Agent-native architecture principles
  • Tool design primitives
  • File organization and context.md
  • Dynamic context injection
  • Execution patterns and model tier selection
  • Self-modification capabilities
  • Testing and product design implications

Use Cases

  • Designing new autonomous agent systems
  • Implementing self-modifying agent applications
  • Building applications where features are agent outcomes
  • Creating MCP tools with atomic primitives

Non-Goals

  • Providing specific code implementations for every pattern
  • Replacing detailed API documentation
  • Dictating a single rigid architectural choice

Practices

  • Agent-native design
  • Tool design
  • System prompt engineering
  • Testing
  • Self-modification

Installation

First, add the marketplace

/plugin marketplace add EveryInc/compound-engineering-plugin
/plugin install compound-engineering@compound-engineering-plugin

Quality Score

Verified
99 /100
Analyzed about 21 hours ago

Trust Signals

Last commitabout 21 hours ago
Stars16.7k
LicenseMIT
Status
View Source

© 2025 SkillRepo · Find the right skill, skip the noise.