Why are XML tags important for Claude Code agents?

XML tags provide clear semantic boundaries that help Claude distinguish between identity, constraints, and knowledge, resulting in more predictable and accurate agent behavior.

Can I use these standards for custom commands?

Absolutely. The skill includes an orchestration schema specifically designed for commands, covering tool permissions, delegation rules, and phase-based execution.

How does the skill handle code blocks inside XML?

It provides specific guidelines for placing code blocks directly under tags and managing character escaping to prevent parsing errors while maintaining code readability.

Does this skill support different agent archetypes?

Yes, it provides specialized XML patterns for Orchestrators, Planners, Implementers, Reviewers, and Testers to ensure each role has the appropriate structure.

Are these standards based on official documentation?

Yes, these patterns are aligned with Anthropic's best practices for prompt engineering and agentic workflow design.

Agent XML Standards

Name: Agent XML Standards
Author: involvex

byinvolvex

0•

Developer Tools

Enforces Anthropic best practices for XML tag structures when designing and implementing specialized Claude Code agents.

The Agent XML Standards skill provides a comprehensive library of structural patterns and schemas designed to optimize Claude Code agents and commands. By implementing standardized tags for roles, instructions, knowledge bases, and formatting, it ensures that custom AI agents follow Anthropic's recommended architectural patterns for maximum reliability. This skill includes specialized structures for various agent types—including Orchestrators, Planners, and Reviewers—providing a robust framework for hierarchical nesting, semantic attribute usage, and consistent error recovery strategies.

Key Features

010 GitHub stars

02Role-specific XML schemas for Orchestrators, Planners, Implementers, and Reviewers

03Strict nesting and semantic attribute rules for structural integrity

04Best practices for character escaping and embedding code blocks within XML

05Standardized core tags for defining agent identity, mission, and expertise

06Pre-defined templates for quality gates, delegation rules, and feedback formats

Use Cases

01Standardizing custom Claude Code commands with consistent workflow patterns

02Building automated code review agents with weighted criteria and specific feedback templates

03Architecting multi-agent systems with clearly defined boundaries and handoffs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add involvex/involvex-claude-marketplace xml-standards

For use in Claude.ai and ChatGPT

Key Features

010 GitHub stars

02Role-specific XML schemas for Orchestrators, Planners, Implementers, and Reviewers

03Strict nesting and semantic attribute rules for structural integrity

04Best practices for character escaping and embedding code blocks within XML

05Standardized core tags for defining agent identity, mission, and expertise

06Pre-defined templates for quality gates, delegation rules, and feedback formats

Use Cases

01Standardizing custom Claude Code commands with consistent workflow patterns

02Building automated code review agents with weighted criteria and specific feedback templates

03Architecting multi-agent systems with clearly defined boundaries and handoffs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add involvex/involvex-claude-marketplace xml-standards

For use in Claude.ai and ChatGPT