think step by step
technique judged 100% likely a named technique or architecture
See how it has spread across media tiers
Mentions over time
Table view
| Week | Articles | Sources |
|---|---|---|
| 2023-W33 | 1 | 1 |
| 2026-W09 | 1 | 1 |
| 2026-W11 | 3 | 1 |
| 2026-W40 | 1 | 1 |
| 2026-W41 | 2 | 2 |
In their words
Prompt Engineering Maturity Model Level Approach Example 1 Ad-hoc prompting "Write a summary" 2 Structured templates "Summarize in 3 bullets: {text}" 3 Few-shot + CoT Examples + "Think step by step" 4 RAG-integrated Context-aware with citations 5 Agent prompts Multi-step with tool use 6 Self-improving Prompt optimization loops Part 1: Foundational Techniques Zero-Shot Prompting # prompting/zero_sh
Think step by step.
It can generate a natural language explanation: "The withdraw function allows an attacker to re-enter the function before the balance is zeroed, draining the contract." More importantly, modern AI tools can propose the fix: function withdraw() external { uint amount = balances[msg.sender]; balances[msg.sender] = 0; // Mitigation: Update state before external call (bool success, ) = msg.sender.call
The Technique Landscape Section titled “The Technique Landscape” | Technique | Core idea | Best for | Cost multiplier | |---|---|---|---| | Zero-shot CoT | ”Think step by step” trigger | Math, logic, multi-step reasoning | ~1.5x (longer output) | | Few-shot CoT | Provide example reasoning chains | Domain-specific reasoning, consistent format | ~2x (examples in prompt) | | Tree-of-Thought | Explore
“Let me think step by step: [reasoning].
This is not the same as prompting a standard model to “think step by step.” When you prompt chain-of-thought, the model is still generating left-to-right in a single pass, just with more verbose output.
A single prompt change that adds “think step by step” to a system prompt can increase output tokens by 3x.
Think step by step and reflect on each response.