← ClaudeAtlas

lean-maxlisted

Max-quality, min-token operating mode for coding and technical work. Use for any code task — bug fixes, features, refactors, reviews, migrations, debugging, multi-file changes — and whenever context is long, tokens are constrained, or work runs in the cloud/headless/CI. Enforces locate-before-read, read-before-edit, verify-before-done, and zero-filler output.
Sarveshmehta1504/lean-max · ★ 1 · AI & Automation · score 72
Install: claude install-skill Sarveshmehta1504/lean-max
# lean-max Two goals, in this order: **the output must be correct**, and **nothing spent that doesn't serve correctness**. Quality is never what you cut. Prose is. Every rule below exists to move tokens *out* of narration and *into* verification. ## 0. Calibrate first (5 seconds, saves the rest) Size the task before choosing a process. Applying heavy discipline to a small task is itself waste. | Signal | Mode | |---|---| | Single file <200 lines, obvious change | **Direct** — read it whole, edit, verify, one-line report. Skip the loop. | | Multi-file, or any file >200 lines, or unfamiliar code | **Full loop** (§1) | | Touches auth, money, migrations, deletes, prod config, or crypto | **Full loop + adversarial pass** — actively try to break your own change before reporting | Never run a heavier mode than the task earns. Never run a lighter one to save tokens. ## 1. The loop 0. **Resume** — if `STATE.md` exists for this project, read it first and say in one line what you're picking up. That file *is* your context; don't re-derive it from the codebase. See `references/context.md`. 1. **Scope** — restate nothing. Name the exact files/symbols in play. If ambiguity would change the work, ask *one* question; otherwise assume, proceed, and note the assumption in one line at the end. 2. **Locate** — grep/glob to the relevant region **and, in the same search, list every dependent** (callers, importers, subclasses, tests, config keys) *before* you edit. Never open a large file t