Behavioral guidelines to reduce common LLM coding mistakes through explicit assumptions, simplicity, and verifiable success criteria.

• Emphasizes surfacing assumptions and tradeoffs upfront rather than making silent decisions or hiding confusion
• Advocates for minimum viable code with no speculative features, abstractions, or error handling beyond what was requested
• Requires surgical, focused edits that touch only what's necessary and match existing code style without improving adjacent code
• Transforms tasks into verifiable goals with explicit success criteria and multi-step plans that enable independent verification loops