# Practice 2: Set Guardrails, Scope & Success Criteria

**Essence**

Draw the box: what’s in scope, what’s out, what must be correct, and what would make the answer unusable.

**What this combines**

* “Set Guardrails” (what’s allowed, what’s not, what gets checked).
* “Set the Bar” (success criteria + red lines).
* Safety/decision emphasis from “Make the Human Call” & “Own the Decision”.

**Why it matters**\
LLMs happily guess. If you don’t separate *creative* from *must-be-true*, you’ll get confident nonsense in the worst places.

**Common traps**

* Not stating any constraints (length, tone, audience, “no fabricated citations”).
* Letting the model improvise on risky domains: numbers, law, policy, internal info.

**Examples**

* “Before you answer, ask me 5 quick questions about audience, length, tone, and any must-include / must-avoid points.”
* “You may be creative with: style/tone/examples. You must *not* be creative with: data, dates, citations, internal policies.”
* “Flag any content that sounds like legal, medical, or financial advice so I can review it.”

***


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://praxistutor.gitbook.io/llm/reframing-the-7-habits-for-reflective-llm-use/section-3-operators-toolkit-practical-examples/7-practices-in-the-operators-toolkit/practice-2-set-guardrails-scope-and-success-criteria.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.