com um clique
cutweight
cutweight contém 5 skills coletadas de brenbuilds1, com cobertura ocupacional por repositório e páginas de detalhe dentro do site.
Skills neste repositório
Session-weight hygiene. Use when sessions run long, when an agent gets slower or starts forgetting mid-session, when deciding between /clear, /compact, or a fresh session, or when every message feels heavier than the work in it. Calls the moment a session has blown the weight limit: cut it right and start the next one light.
Audit the always-on context cost of an agent setup: installed skills, memory files (CLAUDE.md, AGENTS.md), and MCP tool schemas. Use when asked to audit, trim, shrink, or cut skills or context, when an agent picks the wrong skill or feels dumber after the setup grew, when MCP servers eat the context window, or when the token bill grows without more work getting done. Produces a keep/trim/cut table with measured numbers. Does not delete anything.
Cut reasoning-token weight. Use when thinking or reasoning tokens dominate the bill, when asked to cap a thinking budget, lower effort, or stop simple tasks from burning extended thinking. Light tasks get light thinking: matches thinking budget to task weight, then verifies accuracy held before keeping any cap. Never caps blind.
Weigh an agent setup and print a fight-card receipt. Use when asked to weigh in, weigh my agent, check my agent's weight or token overhead, or compare before and after a cut. Measures the always-on context bill (installed skills, memory files, MCP tool schemas), assigns a weight class, prints a shareable card, and shows the delta against the last weigh-in. Read-only except for its own append-only log.
Puts the agent's replies on a word diet. Compressed output that cuts reply tokens by roughly a third to half, more on ultra, while keeping full technical accuracy. Answers like a cornerman between rounds: only what changes the next round. Adds source discipline: verify current-year, recent, latest, changing, risky, or source-sensitive facts before confident answers. Use when user says word-diet, corner mode, cornerman, less tokens, fewer tokens, low tokens, be brief, or /word-diet. Also auto-triggers when token efficiency or fresh source verification is requested.