Skip to content

Commit

Permalink
chore: add system prompt intro
Browse files Browse the repository at this point in the history
  • Loading branch information
niltonheck committed Oct 16, 2024
1 parent 5abd5bf commit 9ee08ad
Show file tree
Hide file tree
Showing 2 changed files with 19 additions and 0 deletions.
4 changes: 4 additions & 0 deletions packages/docs/config/menu/orama-cloud.ts
Original file line number Diff line number Diff line change
Expand Up @@ -86,6 +86,10 @@ const oramaCloudMenu = [
label: 'Custom system prompts',
collapsed: true,
items: [
{
label: 'Introduction to Custom System Prompts',
link: '/cloud/custom-system-prompts/'
},
{
label: 'Customizing the system prompts',
link: '/cloud/custom-system-prompts/customizing-the-system-prompts'
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
---
title: Custom System Prompts
description: Custom System Prompts let you customize AI behavior with tailored instructions, ensuring safe, precise, and efficient responses.
editUrl: false
---

Custom System Prompts in Orama allow you to define custom instructions that shape how the AI engine responds to users. These prompts can be tailored for various use cases, such as altering the tone of the AI, introducing information that isn’t in the data sources, or directing users toward specific actions.

For example, if a user asks about pricing, a system prompt could instruct the AI to suggest visiting a website for details instead of providing a direct answer. You could also include context like software versions or other specialized information to ensure more accurate responses. The flexibility of System Prompts enables you to mold the assistant’s behavior to meet your specific needs.

## Best Practices and Limitations

While System Prompts give you powerful control over the AI’s behavior, they are subject to certain validation rules to maintain safety and efficiency. Orama has implemented jailbreak protection to guard against malicious or harmful actions. There are also limits on the length and complexity of prompts to ensure optimal performance.

These safeguards ensure that while you have the freedom to shape the AI’s responses, the assistant will always behave in a reliable, secure, and efficient manner.

0 comments on commit 9ee08ad

Please sign in to comment.