Skip to content

Prompt Enhancement (Magic Wand) ​

Transform rough, vague prompts into detailed, well-structured queries tailored to your industry and context — with one click.

Overview ​

GhostAI's prompt enhancement feature uses AI to automatically improve your input with the right detail, structure, and industry-specific language. It reads the context around you — your vertical configuration, custom instructions, and recent conversation — to produce a prompt that gets dramatically better results from any model.

How It Works ​

  1. Type a prompt in the chat input (normal chat or comparison mode)
  2. Click the wand icon (✨) next to the submit button, or press Cmd/Ctrl + Shift + E
  3. Watch it transform — enhanced text streams in via a typewriter animation
  4. Undo if needed — click "Undo" within 8 seconds to revert to your original text
  5. Send when ready — review the enhanced prompt and hit Enter

Contextual Awareness ​

Enhancement is not one-size-fits-all. GhostAI incorporates the context available in your session to tailor each enhanced prompt:

  • Industry vertical — Your organization's configured vertical shapes the terminology, compliance framing, and domain expectations applied during enhancement.
  • Custom instructions — Your "About You" and "Response Style" preferences are factored into every enhancement. The more context you provide, the better the results.
  • Conversation thread — Recent messages from your current conversation are considered so the enhancement produces a relevant follow-up rather than a generic standalone prompt.
  • Matter context — When applicable (e.g., legal matters), active case or project context is included for more precise prompts.

Proportional Enhancement ​

The enhancement scales based on your input. Short or vague prompts receive more significant expansion with added structure and constraints, while longer, more detailed prompts receive lighter refinement focused on clarity and precision. This ensures the feature adds value without over-engineering prompts that are already well-formed.

Privacy & PII Safety ​

Enhanced prompts go through the same PII detection and tokenization pipeline as regular messages. No sensitive data is exposed during the enhancement process.

Benefits ​

  • Better responses — Well-structured prompts consistently produce higher-quality AI outputs across all supported models.
  • Domain-appropriate queries — Industry-specific terminology and compliance awareness are built into every enhancement automatically.
  • Faster workflow — Spend less time crafting the perfect prompt and more time working with the results.

Keyboard Shortcut ​

PlatformShortcut
macOS⌘ + Shift + E
Windows/LinuxCtrl + Shift + E

API Endpoint ​

POST /api/chat/enhance

See Chat API Documentation for request/response details.

SOC 2 Type II Certified | BAA Available | HIPAA-Ready