Markdown Token Optimizer

Analyzes markdown files and suggests optimizations to reduce token consumption while maintaining clarity.

When to Use

• Optimize markdown files for token efficiency
• Reduce SKILL.md file size or check for bloat
• Make documentation more concise for AI consumption

Workflow

1. Count - Calculate tokens (~4 chars = 1 token), report totals
2. Scan - Find patterns: emojis, verbosity, duplication, large blocks
3. Suggest - Table with location, issue, fix, savings estimate
4. Summary - Current/potential/savings with top recommendations

See ANTI-PATTERNS.md for detection patterns and OPTIMIZATION-PATTERNS.md for techniques.

Rules

• Suggest only (no auto-modification)
• Preserve clarity in all optimizations
• SKILL.md target: <500 tokens, references: <1000 tokens

Examples

• "Optimize this SKILL.md for tokens"
• "Count tokens in references/SCORING.md"

Troubleshooting

High token count? Check for emojis, repeated headings, or verbose tables.

MCP Tools

CLI fallback: Primary mode is CLI-based. No MCP servers required.

References

• OPTIMIZATION-PATTERNS.md - Optimization techniques
• ANTI-PATTERNS.md - Token-wasting patterns