Success in agentic coding depends on breaking prompts into small, verifiable steps. Andrej Karpathy said in 2026 that "asking for everything at once" fails, and that agents produce consistently correct code when asked in small increments. The core is five rules. ASAP organizes them into copy-paste patterns.

The five rules

Prompting an agent well is the 5 rules Karpathy points to, applied in order to cut the failure rate.

1. Break it small — one PR-sized task at a time; split big features into steps and ask in turn.
2. Give the verification bar — state the done-condition, tests, and expected output in the prompt.
3. Load context first — have it read the relevant files and rules before starting the work.
4. One goal at a time — do not mix several intents into one prompt.
5. Verify-and-feed-back loop — confirm the result with tests and feed failures into the next prompt.

These five are the guardrails against "asking for everything at once."

Why break it small

Asking for a big task at once lets the agent run far without verification. Karpathy said in 2026 that agents moved from "helpful but messy" to "consistently correct code," but that is the story when you ask in small units. When one prompt is too big, there is no point to catch a drift mid-way.

Copy-paste patterns

The rules are best applied as one-line prompt skeletons, the same approach Karpathy uses. The following patterns bake verification in.

The key is that every skeleton carries a verification bar.

Wrap-up

The core of agentic coding prompts is small and verifiable, the lesson Karpathy drew in 2026. Break it small, give the verification bar, load context first, one goal at a time, and verify-and-feed-back. Resisting the urge to ask for everything at once is itself the skill.

Source: ASAP synthesis grounded in Andrej Karpathy, "From Vibe Coding to Agentic Engineering" (Sequoia AI Ascent 2026). Related: agentic coding expertise study (Anthropic, 2026).