PromptisePromptise
Docs
GitHub
Promptise - AI Framework LogoPromptise

The foundation layer for agentic intelligence. Build, secure and operate autonomous AI systems with Promptise Foundry.

pip install promptise

[01] Foundry

  • MCPcast
  • The Promptise Agent
  • Reasoning Engine
  • MCP
  • Agent Runtime
  • Prompt Engineering
  • Execution Engine
  • Agent Identity

[02] Resources

  • Documentation
  • GitHub
  • Guides
  • Learning Paths
  • Questions

[03] Company

  • About
  • Terms of Service
  • Privacy Policy
  • Cookie Policy
  • Subprocessors

© 2026 Promptise by Manser Ventures. All rights reserved.

Open source · Python

← Guides> Library> Topic

System prompts: what to put in them and how to write one

Chat models take messages in roles. The system prompt sets standing rules for the whole conversation: who the assistant is, what it may and may not do, and how it should answer. User messages carry the request of the moment. Models generally give the system prompt more weight, but that is a tendency, not a guarantee, and mixing untrusted text or secrets into it causes problems.

These guides explain what goes in the system prompt and what goes in the user turn, how to set a role or a style guide there, how to keep untrusted input separate from instructions, and best practices for the system prompt of an AI agent that calls tools.

[n.01/02]> 5 guides
  1. 01Role Prompting: When Personas Help and When They Don'tRole prompting explained: when a persona improves output, when it doesn't, and how to write a role that works. Real example: a price-change email.Beginner · 7 minOct 18, 2026
  2. 02System Prompt vs User Prompt: What Goes WhereSystem prompt vs user prompt: what each is for, what to put in each, and how the model weighs them. A real support-assistant example with model output.Beginner · 7 minOct 18, 2026
  3. 03LLM Tone and Style: Write a Style Guide PromptControl LLM tone, voice and length with a style guide in the system prompt: rules, banned words, one example and code checks. Real before-and-after output.Beginner · 6 minOct 18, 2026
  4. 04System Prompt Best Practices for AI Agents in PythonBuild agent system prompts from reusable blocks, switch them by conversation phase, and test them like code. Real Python runs, honest limits.Intermediate · 24 minOct 11, 2026
  5. 05Secure Prompt Design: Separate Instructions From DataSecure prompt design: keep untrusted text as data with message roles, delimiters, spotlighting, explicit rules and output schemas. A real before and after.Intermediate · 9 minOct 18, 2026
[n.02/02]> go further

Related topics

  • Evals and Testing15
  • Output Validation4
  • Prompt Injection13
  • Role Prompting3
  • Structured Outputs8
  • Style and Tone Control3

Topic hubs

  • Building AI agents in Python→
  • Testing and evals→

By category and level

  • AI Engineering guides71
  • Prompt Engineering guides54
  • Beginner guides36
  • Intermediate guides80

Search every guide.

The library lists all guides with filters for topic, level, category and reading time.

All guidesLearning paths