AI Agents Helper - VS Code Plugin

A Visual Studio Code extension developed by Qualex Consulting Services, INC. to assist with AI-powered development workflows.

Engineer: Antonio Martínez (R&D department)

Features

• Seamless integration with Ollama and OpenAI
• Local LLM model management
• Browser-based interface available
• Workspace-aware AI operations

Prerequisites

Before installing the plugin, ensure you have:

1. Python (version ≥3.10) installed

   • Find your interpreter path by running:

     where python
     

2. NodeJS installed

   • Download from: https://nodejs.org/en/download

3. Ollama installed

   • Download from: https://ollama.com/

4. OpenAI API Key

   • Get one at: https://platform.openai.com/docs/overview
     (Note: No credits required)

Installation

1. Install the extension in VS Code:
   • Open Extensions view (Ctrl+Shift+X)
   • Search for "qualex"
   • Install "AI Agents Helper"

Setup

Ollama Configuration

If using Ollama for the first time, download a language model:

ollama pull qwen2.5-coder:7b  # Recommended for low-spec hardware

Alternative recommended models:

ollama pull qwen2.5-coder:14b
ollama pull qwen2.5-coder:32b 
ollama pull mix_77/gemma3-qat-tools:12b

VS Code Configuration

Plugin Settings

Configure the extension through VS Code Plugin settings (Shift+Ctrl+X, or Shift+Cmd+X,):

Navigate to: Extensions → AI Agents Helper Settings

On User and Workspace profiles, add your OPENAI_API_KEY, doesn't need to have credits to work but is neccesary. And add the python path for both profiles the workspace and the user.

   {
     "provider": "ollama",  // or "openai"
     "ollamaModel": "qwen2.5-coder:7b",
     "pythonPath": "/path/to/python",
     "openaiKey": "your-api-key-here"
   }

Note: Find your interpreter path by running:

where python

Installation of Libraries

Open Command Palette (Ctrl+Shift+P/Cmd+Shift+P) and Execute

    >AI Agents: Install QLX Dependencies

Note: If you come from a version previous than 0.15.0 please reinstal libraries. You need to have Ollama running to use ollama as provider.

Usage

Open Command Palette (Ctrl+Shift+P/Cmd+Shift+P) and Execute

    >AI Agents: Show AI Agents Panel

Notes

If changes are made on the settings, close any opened panel of this plugin and restart the service with:

    >AI Agents: Show AI Agents Panel

Prompt Examples

1. Generate code and create a new file:

       Prompt: Create a python function to know if a number is odd and save the code in a file named odd.py
   

2. If you have two files in your workspace, let's say dataset.py and hello.py, use them as context to improve code

       #dataset.py
       import pandas as pd
       mydata = pd.DataFrame({
           'id': ['1', '2', '3'],
           'age': [25, 30, 28],
           'salary': [50000, 60000, 55000]
       })
   

       #hello.py
       import pandas as pd
       from .data import mydata
   
       # Add a new variable called 'bonus' and assign a value based on salary
       mydata['bonus'] = mydata.apply(lambda row: 5000 if row['salary'] > 55000 else 2500, axis=1)
   
       # Print the data set
       print(mydata)
   

       Prompt: Given dataset.py and hello.py add the column of years of experience, find the correlation and create a graph.
   

3. Code conversion:

       Prompt: convert from sas to python the file example_sas.sas
   

4. AWS S3 Bucket Operations:

       Prompt: list my bucket my-bucket-name
   

       Prompt: upload file.txt to my-bucket-name and then list the content
   

       Prompt: List all PDF files in bucket my-company-data under prefix 'documents/'
   

       Prompt: Download all files from bucket 'backup-bucket' with prefix 'user-uploads/john/' to local './tmp/john_uploads'
   

       Prompt: Delete all files in bucket 'temp-uploads' that are older than 30 days. First list them for confirmation.