• name follows naming rules (lowercase, hyphens, 1-64 chars)
• name matches parent directory name
• name uses gerund form when possible (processing-pdfs)
• description is third person ("Extracts...", not "I extract...")
• description includes what it does AND when to use it
• description has no markdown formatting
• description includes trigger keywords for discovery
• SKILL.md body is under 500 lines
• Additional details are in separate reference files
• No time-sensitive information (or in "old patterns" section)
• Consistent terminology throughout
• Examples are concrete, not abstract
• File references are one level deep from SKILL.md
• Reference files >100 lines have table of contents
• Progressive disclosure used appropriately
• Workflows have clear steps

Scripts (if applicable)

• Scripts are tested by actually running them (not just written)
• Scripts solve problems rather than punt to agent
• Error handling is explicit and helpful
• No magic constants (all values justified/documented)
• Required packages listed and verified as available
• Scripts have clear documentation
• No Windows-style paths (use forward slashes)
• Validation/verification steps for critical operations
• Feedback loops included for quality-critical tasks