Skip to main content

​
What can you do with it?

Use Gemini for text processing, analysis, and multimodal tasks. Perfect for complex reasoning, content generation, and file analysis. Supports both synchronous processing for quick tasks and asynchronous processing for heavy workloads with large files.

​
How to use it?

​
Basic Command Structure

/gemini [prompt] [optional-parameters]

​
Parameters

Required:
  • prompt - Your instructions or questions for Gemini
Optional:
  • model - Specific Gemini model to use (defaults to gemini-2.5-flash)
  • system prompt - Override the default system prompt
  • files - File URLs to include (supports images, documents, and text files)
  • async - Set to true to force asynchronous processing for heavy tasks

​
Response Format

Synchronous Response (light tasks): 
{
  "output": "Gemini's generated response",
  "metadata": {
    "model": "gemini-2.5-flash",
    "tokens": {
      "input": 150,
      "output": 300
    }
  }
}
Asynchronous Response (heavy tasks): 
{
  "responseId": "job-12345",
  "status": "queued",
  "createdAt": "2025-10-08T16:08:48.616Z",
  "message": "Process started. Once completed, you can see the result in the file storage collection",
  "outputFileName": "result.txt",
  "placeholderFile": {
    "fileId": "abc123",
    "signedUrl": "https://skills.pinkfish.ai/files/collection/result.txt",
    "fileName": "result.txt",
    "collectionId": "collection-id",
    "contentType": "text/plain"
  }
}

​
Examples

​
Synchronous Usage (Light Tasks)

/gemini
prompt: Explain the concept of machine learning in simple terms
Gets an immediate response from Gemini for quick text processing. 
/gemini
prompt: Analyze this document and extract key points
files: document.pdf
Processes small files synchronously with immediate results.

​
Asynchronous Usage (Heavy Tasks)

/gemini
prompt: Generate a comprehensive analysis of this large dataset
files: large_document.pdf
async: true
For complex analysis that takes time - returns job tracking info and processes in background. 
/gemini
prompt: Create a detailed summary of these multiple documents
files: doc1.pdf, doc2.pdf, doc3.pdf
async: true
Multiple large files automatically use async processing with placeholder file for results.

​
Notes

Processing Options:
  • Sync: Quick tasks with immediate results - best for text processing and small files
  • Async: Heavy tasks with background processing - returns a placeholder file URL that you can check to monitor progress and see results when complete
  • File Support: Images, documents, PDFs, and text files
  • Output: Results saved as text files with markdown summaries for async tasks
  • Monitoring: Check the placeholderFile.signedUrl to see processing status and final results
For specialized tasks: See LLM File Type Support for detailed information about file formats supported by Gemini and other models.

​
Supported Models

Choose the appropriate Gemini model based on your specific needs:
  • gemini-2.5-flash (default) - Cost-efficient model for fast, general tasks
  • gemini-2.5-pro - Enhanced thinking and reasoning, multimodal understanding, advanced coding
  • gemini-2.5-flash-lite - Cost efficiency and low latency for cost-sensitive, basic tasks
  • gemini-2.0-flash - Next generation features with speed and thinking for high-volume tasks
All models support images, documents, and text processing.