Use when writing instructions that guide Claude behavior - skills, CLAUDE.md files, agent prompts, system prompts. Covers token efficiency, compliance techniques, and discovery optimization.
1. Claude is smart. Only write what it doesn't already know. Challenge each line: does this justify its token cost?
2. Positive > Negative framing. "Don't do X" triggers thinking about X (pink elephant problem). Say what TO do, not what to avoid.
# Bad: triggers the behavior
Don't create duplicate files
# Good: directs to correct behavior
Update existing files in place
3. Context motivates compliance. Explain WHY, not just WHAT. Claude generalizes from motivation.
# Less effective
NEVER use ellipses
# More effective
Your response will be read aloud by a text-to-speech engine, so never use ellipses since the TTS engine cannot pronounce them.
4. Placement matters. Instructions at prompt start and end receive higher attention. Critical rules go at boundaries.
5. ~150 instruction limit. More instructions = uniform degradation across ALL rules. Prune ruthlessly.
6. Repetition enforces critical rules. For high-stakes requirements, repeat with different framings.
Targets:
Progressive disclosure: Main file is overview + links. Reference files load on-demand.
The description field determines if Claude finds your skill.
Format: Start with "Use when..." + specific triggers + what it does.
Write in third person. Injected into system prompt.
# Bad: vague, first person