Meerkat · Field guide No. 01open the builder →
§ field guide

Six ways to ask
a machine.

  1. 01Direct
  2. 02Structured
  3. 03Few-shot
  4. 04Chain-of-thought
  5. 05Just-in-time
  6. 06Sub-agent
01 / DIRECT

Direct. One sentence in.

Most people over-engineer. The model is smarter than the average prompt assumes. Direct is what you reach for when the task is obvious to a thoughtful human and the format doesn't matter much. Strip away the role-playing and the ten bullet rules, ask plainly, and accept that the floor is already pretty good.

reach for it when
  • +Tasks the model already does well by default
  • +Quick edits, rewrites, simple lookups
  • +When you trust the model's defaults more than your own scaffolding
avoid it when
  • Anything with format requirements
  • Multi-step reasoning
  • High-stakes output you can't eyeball
02 / RTCR

Structured. Role · Task · Context · Rules.

If a senior editor handed you a brief, it would have four parts: who's writing, what they're writing, who it's for, and what not to do. Structured prompts are that brief. Boring? Slightly. Reliable? Extremely. This is the default for a reason — it's the prompt you'd want a junior to write before they touched the keyboard.

reach for it when
  • +Most production work
  • +Anything you'll run more than once
  • +Output that has to land a specific way
avoid it when
  • Throwaway one-shot questions
  • Open-ended brainstorms where rules suppress range
03 / EXAMPLES
❍❍

Few-shot. Show, don't tell.

Rules tell the model what to do. Examples show it. Two well-chosen pairs (input → desired output) outperform a wall of guidelines almost every time, because language models are pattern matchers wearing a reasoning costume. The hard part isn't the prompt — it's curating examples that aren't subtly contradictory.

reach for it when
  • +Output formats hard to describe in words (tone, structure, voice)
  • +Classification and labeling
  • +Anything where the gold standard is easier to demonstrate than define
avoid it when
  • When you don't have good examples yet
  • Tasks where examples would prime a single answer (creative work)
04 / REASONING
◐→●

Chain-of-thought. Make the model think out loud before it commits.

When you ask a person a hard question, the good ones pause. Models can pause too — if you let them. A chain-of-thought prompt explicitly instructs the model to reason through the problem in writing before stating a conclusion. The output is longer, slower, and noticeably better on anything that requires more than recall.

reach for it when
  • +Multi-step reasoning, math, logic
  • +Decisions that need a defensible 'why'
  • +Catching mistakes before they get baked into the answer
avoid it when
  • Latency-sensitive UX
  • Short factual lookups (you'll get a lecture)
05 / RETRIEVAL

Just-in-time. Don't stuff the context.

Stuffing every possible document into the prompt is the AI equivalent of bringing a library to a coffee meeting. Just-in-time prompts assume the model is an agent with tools, and instruct it to retrieve information when it needs it — not before. The prompt becomes a procedure, not a payload. (Inspired by Anthropic's writing on context engineering for agents.)

reach for it when
  • +Agents with tools (search, file read, code exec)
  • +Long-running tasks where the full context won't fit
  • +When source-of-truth data is large and changing
avoid it when
  • Plain chat completions with no tool access
  • When you actually do have all the context handy
06 / ORCHESTRATE

Sub-agent. One prompt to plan.

A single prompt trying to do everything is a single prompt failing at everything. The sub-agent pattern splits the work: a planner prompt decomposes the job into discrete sub-tasks, each handed off to a fresh, focused prompt with only the context it needs. The planner reassembles the results. Slower per-call, dramatically more reliable end-to-end.

reach for it when
  • +Long, multi-step jobs (research, document generation, codebase changes)
  • +Tasks that exceed a single context window
  • +When you want auditable intermediate steps
avoid it when
  • Anything you can solve in one round-trip
  • Tight latency budgets (sub-agents add overhead)
still reading?

The right approach is whichever ships.

Pick a job, pick an approach, watch Meerkat assemble the prompt in front of you. You can change your mind on the next click.