PromptCorrectlyPromptCorrectly
StudioAI CoursesLibraryBlogPricingAbout
Log inStart free
PromptCorrectlyGLOSSARY · SYSTEMS & SAFETY
Home/Glossary/Guardrails

What is Guardrails?

Guardrails are the rules and checks around a model's behaviour — what it refuses, what it must include, and validators that catch bad output before it reaches a user.

Guardrails live in three places: in the prompt (hard rules and refusals), in code (validators that check format, banned content or missing fields), and in process (a human reviews anything high-stakes). The prompt alone is not a guardrail — models can be talked out of instructions.

For personal use, a guardrail is simply a rule you never drop: "cite the passage", "flag uncertainty", "don't write the essay for me". For products, it's the difference between a demo and something you can ship.

How to use it well
  1. State the rule, the exact behaviour when it's triggered, and the fallback.
  2. Validate structured output in code; don't trust the model to self-report.
  3. Keep humans in the loop for money, health, legal and safety.
Go deeper
Prompt Evaluation Rubrics (course)
Related terms
ConstraintsPrompt injectionHallucinationSystem promptAll terms →

Use it right now

Ask our brain anything on the homepage — it remembers the whole conversation — or write a brief in the Studio and see the prompt it compiles to.

Ask the brainOpen the Studio
promptcorrectly.comFROM VAGUE INTUITION TO STRUCTURED INSIGHTBYOK · ANTHROPIC · OPENAI · GROK
PromptCorrectlyPromptCorrectly

The visual workspace for people who actually use AI. Built in the open, priced for humans, powered by your own keys.

Product
  • Studio
  • AI Courses
  • Library
  • Pricing
Resources
  • How it works
  • Templates
  • Community library
  • Repository
  • All 3,600 prompts
  • How to prompt correctly
  • Prompts for every field
  • AI glossary
Company
  • About
  • Contact
  • FAQ
  • Pricing
© 2026 PromptCorrectly · From vague intuition to structured insight.
TermsPrivacyRefundDisclaimerAcceptable useCookies