save_prompt
Save a new prompt to your Promptsy vault directly from the conversation.
Overview
The save_prompt tool creates a new private prompt in your vault. It's perfect for capturing prompts you've refined during a conversation without switching apps.
Parameters
| Parameter | Type | Required | Max Length | Description |
|---|---|---|---|---|
title | string | Yes | 200 chars | Prompt title |
body | string | Yes | - | Full prompt text |
description | string | No | 500 chars | Brief description |
category | string | No | - | One of the predefined categories |
models | string[] | No | 20 items | AI models this works with |
use_cases | string[] | No | 10 items | Scenarios for using this prompt |
tag_names | string[] | No | 10 items | Tags (created if new) |
Categories
Valid category values:
generalwritingcodinganalysiscreativebusinesseducationproductivitycommunicationother
Usage Examples
Basic Save
"Save this prompt as 'Email Summarizer'"
{
"tool": "save_prompt",
"parameters": {
"title": "Email Summarizer",
"body": "Summarize the following email in 3 bullet points..."
}
}
With Metadata
"Save this as a coding prompt called 'Code Reviewer' with tags 'development' and 'quality'"
{
"tool": "save_prompt",
"parameters": {
"title": "Code Reviewer",
"body": "Review the following code for bugs and best practices...",
"category": "coding",
"models": ["Claude", "GPT-4"],
"use_cases": ["Pull request reviews", "Code audits"],
"tag_names": ["development", "quality"]
}
}
Full Example
{
"tool": "save_prompt",
"parameters": {
"title": "Technical Blog Post Writer",
"body": "Write a technical blog post about {{topic}}. The post should be approximately {{length}} words...",
"description": "Generates well-structured technical blog posts with code examples",
"category": "writing",
"models": ["Claude", "GPT-4", "Gemini"],
"use_cases": ["Content marketing", "Developer documentation", "Tutorial creation"],
"tag_names": ["content", "blogging", "technical-writing"]
}
}
Response
Text Response
A confirmation message with a link to the prompt:
"Saved 'Code Reviewer' to your vault. Edit in Promptsy →"
Structured Content
{
prompt: {
id: string;
title: string;
body: string;
description: string | null;
category: string | null;
models: string[];
use_cases: string[];
tags: { id: string; name: string }[];
is_shared: false; // Always private
created_at: string;
},
url: string // Direct link to edit the prompt
}
Widget Display
After saving, a confirmation widget is shown with:
- Success message
- Prompt title
- Quick link to edit in Promptsy
- Summary of metadata added
Tier Limits
Free Tier
- Maximum 25 private prompts
- If limit reached, you'll get a
LIMIT_REACHEDerror
Solo & Team Tiers
- Unlimited private prompts
Errors
| Error Code | Description | Solution |
|---|---|---|
AUTH_REQUIRED | Not authenticated | Authorize the integration |
INSUFFICIENT_SCOPE | Missing prompts:write scope | Re-authorize with correct scopes |
LIMIT_REACHED | Hit 25 prompt limit (Free tier) | Upgrade to Solo or delete prompts |
VALIDATION_ERROR | Invalid parameters | Check parameter formats |
Tips
Using Variables
Include {{variable}} syntax for reusable prompts:
"Save a prompt for code review with a variable for the programming language"
Review the following {{language}} code for:
- Bugs and potential issues
- Best practices
- Performance improvements
Auto-Tagging
Tags are created automatically if they don't exist:
"Save this with tags 'new-tag' and 'another-tag'"
Both tags will be created and assigned to the prompt.
Bulk Context
When saving a refined prompt from a conversation, ask:
"Save our final version of the email prompt with all the improvements we discussed"
The AI will capture the complete prompt including any modifications.
Related Tools
- browse_prompts - View your saved prompts
- get_prompt - View full content of a prompt
- search_public_prompts - Find community prompts
- fork_prompt - Fork public prompts to your vault