Author

aiterms

What is a system prompt?

A system prompt is the instruction text that sets the assistant’s role, behavior, and boundaries before the user’s message is processed.

Why it matters

System prompts help you control how an LLM behaves without changing the model itself. They’re useful when you want consistent tone, safety rules, formatting, or task-specific behavior across many requests.

In practice, teams use system prompts to:

• keep outputs on-brand,
• enforce policies or guardrails,
• define what the assistant should and should not do,
• reduce ambiguity before the user’s instructions arrive.

If you are building an AI product, the system prompt is often the first place to encode “how this assistant should behave.”

How it works

Most chat-style LLM APIs accept multiple message roles, and the system prompt is the highest-priority instruction in that conversation structure. It is typically provided separately from the user’s message, so the model can treat it as setup rather than as part of the user request.

The exact semantics depend on the model and API, but the general idea is stable: the system prompt establishes global instructions for the session or request. User prompts then ask for a specific task within those boundaries.

A good system prompt is usually short, explicit, and behavior-focused. It should describe the assistant’s job, constraints, and any important style rules. It should not try to cram in every possible edge case; that tends to become brittle.

Tiny concrete example

System: You are a concise support assistant. Answer in plain English. If you are unsure, say so.

User: Can you explain why my export failed?

Assistant: I can help. Common causes are missing permissions, an invalid file path, or a timeout. If you share the error message, I can narrow it down.

Here, the system prompt sets the tone and uncertainty policy. The user asks the question, but the answer should still follow those instructions.

Common pitfalls / when NOT to use it

• Don’t use it for user-specific content. System prompts are for behavior and policy, not for data that changes per user or per request.
• Don’t overload it. If it becomes a giant rules document, the model may follow it inconsistently and your app becomes harder to maintain.
• Don’t assume it is a security boundary. It helps steer behavior, but it is not a guarantee against prompt injection or misuse.
• Don’t hide product logic only in the prompt. If a rule is critical, enforce it in application code too.
• Don’t confuse it with fine-tuning. A system prompt changes instructions at inference time; fine-tuning changes the model’s learned behavior.

Related terms

• What is prompt engineering?
• What is few-shot prompting?
• What is zero-shot prompting?
• What is chain-of-thought prompting?
• What is context engineering?
• What is a context window?
• What is prompt caching?
• What is structured output (JSON mode)?
• What is prompt injection?

同じ著者の記事

What is prompt engineering?

Prompt engineering is the practice of designing and refining the text you give to a language model so it produces more useful, reliable, and specific outputs. Large language models are sensitive to how you ask. A vague request can produce a vague answer; a well-structured prompt can improve format, scope, tone, and task performance. You reach for prompt engineering when you want to: get a model to follow instructions more consistently, steer it toward a role, style, or output format, reduce back

papoo.work

What is contextual / late chunking?

Contextual chunking, also called late chunking, is a way to split a document into retrieval chunks after a model has already encoded the whole document, so each chunk keeps some awareness of the surrounding context. Normal chunking cuts text first and then embeds each piece alone. That is simple, but it can make chunks semantically thin: a paragraph may be ambiguous without the surrounding sections, headings, or earlier definitions. Late chunking helps when you want better retrieval quality

papoo.work

What is a knowledge base for RAG?

A knowledge base for RAG is the collection of documents or records a system retrieves from to ground an LLM’s answer in relevant source material. RAG, short for retrieval-augmented generation, works best when the model can search a useful, trustworthy set of content before it writes. That content is the knowledge base. You’d reach for one when the model needs to answer questions from: company docs support articles policies and handbooks product specs research notes tickets or tickets-like record

papoo.work

What is top-k retrieval?

Top-k retrieval is a way to search a collection and return the k most relevant items for a query, instead of trying to sort or inspect everything. Top-k retrieval is the standard first step in search, recommendation, and retrieval-augmented generation (RAG). You use it when you want a fast shortlist: the most relevant documents, passages, products, images, or candidates, without paying the cost of full ranking over the entire corpus. In practice, teams reach for top-k retrieval when: the dat

papoo.work

What is a reranker / cross-encoder?

A reranker, often implemented as a cross-encoder, is a model that takes a query and a candidate result together and scores how well they match. A reranker solves the “good enough first pass, better final answer” problem. In search and retrieval systems, you usually start with a fast retriever that finds a few dozen or few hundred candidate documents. That first stage is optimized for speed, not perfect ranking. A reranker then looks at those candidates more carefully and reorders them so the mos

papoo.work

What is hybrid search (BM25 + vector)?

Hybrid search is a retrieval method that combines keyword search with vector similarity search so you can find documents that match both the exact words a user typed and the broader meaning of the query. Pure keyword search is good when the exact term matters. Pure vector search is good when wording varies but meaning is similar. In practice, many real queries need both. You reach for hybrid search when: users may search with exact product names, error codes, or technical terms the same

papoo.work

What is retrieval-augmented generation (RAG)?

Retrieval-augmented generation, or RAG, is a way to make a language model answer questions using relevant external documents it retrieves first, instead of relying only on what it memorized during training. RAG helps when you want answers that are: grounded in your own content, such as docs, policies, tickets, or product knowledge easier to update than retraining a model more likely to mention current or proprietary information In practice, teams reach for RAG when a model needs to answer “what

papoo.work

What is a vector database?

A vector database, also called a vector store, is a system for storing and searching data by its numerical embeddings so you can find items that are “most similar” rather than exactly matching text or IDs. Regular databases are great when you know the exact thing you want: a user ID, a product SKU, a timestamp range. But many AI features need semantic search or nearest-neighbor retrieval: “find questions like this one,” “retrieve passages relevant to this prompt,” or “recommend items

papoo.work

What is an embedding?

An embedding is a vector embedding: a list of numbers that represents something—like a word, sentence, image, or user—in a way a model can compare mathematically. Embeddings turn messy, human data into a form machines can use for: Similarity search: “Find documents like this one.” Recommendations: “Show items close to what the user liked.” Clustering and classification: “Group related things together.” Retrieval for LLMs: “Fetch the most relevant context before generating an

papoo.work

What is semantic search?

Semantic search is a way to find information by meaning, not just by matching the exact words you typed. Traditional keyword search is good when you know the exact term to look for. Semantic search helps when users phrase the same idea in different words, use synonyms, or ask a question instead of naming a document. You’d reach for it when: search queries are messy or conversational, exact wording varies a lot across documents, you want to retrieve “about the same thing” rather than “contains th

papoo.work