Command Management Standards

Shared Standards

For common principles, integration patterns, and anti-patterns, see:

• shared-standards.md - Common standards for all management skills

Essential Principles

1. Wrapper Commands are Anti-Patterns: If a command only delegates to a single skill/agent, delete the command and configure the skill with user-invocable: true and context: fork.
2. Commands Orchestrate Multi-Skill Workflows: Commands exist to sequence multiple Skills or provide complex delegation patterns.
3. User-Centric Design: Commands are convenient shortcuts for human users, not programmatic interfaces.

Capability Index

Command Standards

File: command-standards.md Use for: Creating new slash commands, selecting templates, semantic categorization, YAML frontmatter, security patterns.

Covers:

• Command structure standards
• Semantic categories (Verbs, Personas, Objects, Execution)
• Template selection criteria
• YAML frontmatter requirements
• User-centric wrapper patterns
• Dynamic context standards
• Security and tool restrictions
• Validation and testing protocols

Additional References

Reference	Purpose
syntax-guide.md	Arguments, Frontmatter, and Categories
background-execution.md	Background execution patterns
permissions-guide.md	Security and tool restrictions
cross-platform.md	OS-specific considerations

Asset Library

Templates (assets/templates/)

[!CAUTION] Law 2 Enforcement: Single-skill wrappers are glue code. Before creating a Command, verify it orchestrates 2+ distinct Skills. If it only wraps one Skill, configure that Skill with user-invocable: true and context: fork instead.

Deprecated Templates (Moved to .attic):

• context-loader.md - Dynamic context loading is now handled by context: fork Skills
• forked-skill.md - Skill binding is achieved via agent: [name] in Skill frontmatter

Removed Anti-Pattern Templates:

• skill-delegator_*.md - Single-skill wrappers are glue code. Use user-invocable: true on the skill instead.
• agent-delegator.md - Single-agent wrappers are glue code. Use context: fork with agent: [name] on the skill instead.

Examples (examples/)

• bash-logic.md - Git commits and deployment gates with dynamic context

Anti-Patterns

Pattern	Why Avoid	Alternative
Vague description	Won't discover	Add specific purpose
XML in description	Law 4 violation	Natural Language + USE triggers
Missing tool restrictions	Security risk	Add allowed-tools
No dynamic context	Missing state	Add ! context
Poor argument integration	Arguments ignored	Use $ARGUMENTS
Overly complex	Hard to use	Split into multiple commands
Interactive prompts	Breaks async	Make autonomous

Description Standards (Law 4)

• NO XML: Do not use <example> tags in frontmatter.
• Natural Language: Use "USE when [condition]" sentences.
• Keywords: Include a Keywords: word1, word2 line for fuzzy matching.

Validation Protocol

Any command MUST pass:

• Valid YAML frontmatter
• Clear, specific description
• Semantic category selected
• Proper naming conventions
• Arguments properly handled
• Tool restrictions added (if needed)
• Dynamic context included (if needed)
• disable-model-invocation set (if wrapper)
• Tested with real invocation

Best Practices

See references/command-standards.md for comprehensive best practices including:

• Command naming conventions
• Argument handling patterns
• Description standards
• Template selection logic

Continuous Improvement

Refactoring Triggers

Consider refactoring when:

• Command becomes too complex
• Usage patterns change
• New patterns emerge
• Standards evolve

Refactoring Principles

• Enforce clean breaks over backward compatibility (Codebase > History)
• Preserve core functionality
• Improve clarity and usability
• Reduce complexity
• Enhance maintainability