agent-native-reviewer

Audit application codebases for gaps between user-accessible features and agent-capable actions.

1|Updated Apr 4, 2026
One-click install
npx skills add https://github.com/ybbms777/compound-engineering --skill agent-native-reviewer-ybbms777
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-native-reviewer
Source: https://github.com/ybbms777/compound-engineering/tree/main/skills/agent-agent-native-reviewer
Command: npx skills add https://github.com/ybbms777/compound-engineering --skill agent-native-reviewer-ybbms777

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the problem of applications having features only accessible to human users, leaving AI agents unable to perform the same actions and creating inconsistent experiences for end users interacting via agents.

Core Features & Use Cases

  • Action Parity Validation: Identifies UI actions (buttons, forms, navigation) with no equivalent agent tool, ensuring agents can complete all core user workflows.
  • Context Parity Checks: Verifies agents have access to the same runtime data, domain vocabulary, and resource information as human users.
  • Anti-Pattern Detection: Flags common agent-native design flaws including workflow tools that embed business logic, sandbox isolation of agent data, and context starvation in system prompts.
  • Use Case: Ideal for teams building AI-integrated web apps, CLI tools, or platforms that want to avoid orphan features and treat agents as first-class citizens.

Quick Start

Use the agent-native-reviewer skill to audit your latest code changes for gaps between user-accessible features and agent-capable actions.

Frequently Asked Questions about agent-native-reviewer

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

FAQPage Schema
What is agent-user feature parity in AI-integrated web apps?

Agent-user feature parity ensures AI agents can perform the same actions and access the same context as human users. This Skill audits codebases to identify gaps where UI features lack equivalent agent tools, preventing inconsistent application experiences.

How do I check my codebase for missing agent tools?

To check your codebase for missing agent tools, use this Skill to validate action parity. It scans UI components like buttons and forms, then flags workflows lacking equivalent agent-accessible tools to ensure first-class AI citizen architecture.

Does this tool detect agent-native design anti-patterns like context starvation?

Yes, this tool detects agent-native design anti-patterns including context starvation in system prompts, sandbox isolation of agent data, and workflow tools embedding business logic. It reviews your architecture to ensure agents receive adequate runtime data.

Can I audit CLI tools for action parity between users and AI agents?

Yes, you can audit CLI tools for action parity. This Skill evaluates AI-integrated platforms and CLI tools to verify shared workspace structure, context parity, and primitive tool design, ensuring equivalent capabilities for both human users and agents.

Why does my AI agent fail to complete core user workflows?

Your AI agent may fail user workflows due to orphan features lacking equivalent agent tools. This Skill identifies action parity gaps where human-accessible UI navigation and forms have no corresponding primitive tool design for agents.

What is the best way to validate agent-native architecture design patterns?

The best way to validate agent-native architecture design patterns is auditing action and context parity. This Skill reviews primitive tool design and shared workspace structures to ensure AI agents possess equivalent capabilities and domain vocabulary access.