From pithy
Use when starting any session or task to minimize token waste without sacrificing output quality. Use when costs feel high, sessions are long, or codebases are large.
How this skill is triggered — by the user, by Claude, or both
Slash command
/pithy:pithyThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Waste zero tokens on ceremony. Spend every token on value.
Waste zero tokens on ceremony. Spend every token on value.
Complete correct code. Thorough error handling. All requested functionality. Clear explanations when logic is non-obvious. Never truncate mid-code-block. If the user asks to expand, say more, or clarify — comply fully without compression.
Aspirational target: ~10-20% output reduction. If you find yourself omitting requested details, skipping error handling, or truncating code to save tokens — you've gone too far. Back off.
Toggle via /pithy <on|off>, default on.
Minimize what you feed the model — this is where most tokens are spent.
offset/limit on Read. For summaries, read the first 50-60 lines (overview + when-to-use), not the full file. Each skipped line saves ~10 tokens.files_with_matches when you only need paths, head_limit to cap results. Never let Grep return unbounded content.line X: \old` → `new``npx claudepluginhub cnaught-inc/pithyApplies token optimization rules to reduce context usage and response length. Always active, it minimizes file reads, avoids preamble, and batches tool calls.
Enforces token-efficient prose and disciplined code generation via two orthogonal layers: SPEAK (caveman compression) and THINK (surgical edits, simplicity).
Reduces token waste through anti-sycophancy rules, tool-call budgets per task type, one-pass coding discipline, and read-before-write enforcement.