← ClaudeAtlas

complexity-guardlisted

Applied Reality Engineering — The Rober Rules. Diagnostic counterpart to Murphy's Law. Auto-invokes when work drifts, stalls, or spirals.
rudi193-cmd/willow-2.0 · ★ 2 · AI & Automation · score 59
Install: claude install-skill rudi193-cmd/willow-2.0
# The Rober Rules — Applied Reality Engineering Named for Mark Rober. The diagnostic counterpart to Murphy's Law. Murphy tells you what will go wrong. The Rober Rules tell you what to do about it. **Source:** Prof. Penny Riggs / UTETY | Lineage: K.I.S.S. methodology, three-layer recursion limit (AIONIC), next-bite methodology, Bartlett (1958) point-of-no-return. --- ## The Rules 1. If you think you have gone too far, you have. Turn back. 2. The exit must be smaller than the system. 3. Three steps off the path, return to the path. 4. Test one thing. Learn. Then take the next bite. 5. We do not guess. We measure, or we test. 6. Failure is data. Listen to it. 7. Build it simple enough that stupidity cannot break it. 8. When the jig needs a jig, start over. 9. The doubt is the data. 10. *(Apocryphal)* The 10mm socket is already gone. --- ## When to Apply **Rule 1** — You've gone too deep into a rabbit hole. Stop. Back out. **Rule 3** — You're three tool calls from the original ask. Return to first principles. **Rule 4** — Don't batch-fix everything at once. One change. Verify. Next. **Rule 5** — You're assuming the cause of an error. Don't. Check. **Rule 6** — A failing test or hook is telling you something. Read it. **Rule 7** — If the implementation needs explaining, it's too complex. **Rule 8** — If the abstraction requires another abstraction, flatten and restart. **Rule 9** — The moment you feel uncertain, that uncertainty is valid data. ΔΣ=42