Instruction
An instruction is a directive that specifies an action, task, operation, constraint, or expected behavior for a person or computational system.
Prompt Engineering in Practice
An instruction is a directive within a prompt that specifies the task the model should perform.
- How it is used
- Defines the task within a structured prompt for software-engineering work, from incident updates to deployment decisions.
- How deep it goes
- Defines instructions and gives writing rules for actions, audience, scope, and sequence, without descending into model internals.
- Examples
- Uses secure logging guidelines, an incident status update, and a deployment GO/NO-GO brief as structured prompt examples.
- What you can do after
- Build it from scratch by drafting a clear task directive, separating other prompt elements, and placing stable instructions for repeated use.
Read how Prompt Engineering in Practice explains instruction →
Building LLM Applications with DSPy
Text within a prompt that directs a language model’s behavior or specifies how it should perform a task.
- How it is used
- Directs language-model behavior in a DSPy prompt, including how to distinguish intent labels.
- How deep it goes
- Explains placement, defaults, inspection, manual modification, and optimization, without descending into language-model internals.
- Examples
- Uses flight-intent label definitions, instruction inspection, and a Python module that appends a closing direction.
- What you can do after
- Operate/configure it by writing and inspecting signature instructions, or optimize it alongside demonstrations with MIPROv2.
Read how Building LLM Applications with DSPy explains instruction →
Coding with AI
A directive that specifies an action or response an artificial-intelligence model should produce.
- How it is used
- Uses directives to turn an LLM interaction from a search-like query into a requested action or response.
- How deep it goes
- Defines clarity, directness, simple language, and task breakdown, but does not descend into model internals.
- Examples
- Uses existing project context, concrete examples, conversational clarification, and stepwise task requests.
- What you can do after
- Apply it to write clearer prompts and improve model responses through clarification and task decomposition.
LLMs in Production
The instruction component of a prompt is the task-specific directive that specifies what the language model should do.
- How it is used
- Directs a language model toward a task-specific response within a structured prompt or reusable template.
- How deep it goes
- Defines the instruction’s role and illustrates it with Q&A and audience-level examples, without examining model internals.
- Examples
- Uses a 401k Q&A template and an instruction that explains the answer as if addressing a five-year-old.
- What you can do after
- Apply it via a prompt template to specify a model’s task and desired response behavior.
The Art of AI Product Development
A prompt instruction is the task specification that tells a model what operation to perform.
- How it is used
- Specifies the operation a language model should perform within a prompt.
- How deep it goes
- Defines instructions and places them among prompt components, stopping short of implementation details.
- Examples
- Uses an EU AI Act article request and reusable topic, content-type, context, and demonstration patterns.
- What you can do after
- Apply via reusable prompt templates to specify tasks, audiences, topics, qualities, and supporting information.
Read how The Art of AI Product Development explains instruction →
Related concepts
- constraintcontrasts with · A constraint is a condition or rule that restricts the permissible states, values, actions, or solutions in a system or problem.
- promptpart of · An input, typically containing instructions or context, supplied to an artificial-intelligence model to guide or elicit its output.
- system promptincludes · A system prompt is a high-priority instruction or context supplied to an AI model to establish its role, task, behavioral guidance, or constraints.
- agent skillincludes · An agent skill is a reusable package of instructions, capabilities, or supporting resources that guides an AI agent in performing a task or workflow.
- COPROconfigured by · COPRO is a DSPy prompt-optimization technique that iteratively generates and evaluates alternative prompt instructions to improve their performance.
- MIPROv2produced by · MIPROv2 is a DSPy optimizer that searches for effective instructions and demonstrations for language-model programs using performance feedback and a Bayesian proposal model.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.