Demo Context Aware Terminal Suggestions

In this lesson we demonstrate how context-aware terminal suggestions speed up development by understanding your project structure, file contents, and environment. With full context, an assistant can generate accurate terminal commands, create tests, and help troubleshoot failures.

A presentation slide reading "Context-Aware Terminal Suggestions" with a large dark-blue curved shape on the right containing the word "Demo." The bottom-left corner shows a small "© Copyright KodeKloud" label.

Cursor AI already translates natural language into terminal commands. Its usefulness increases when it can read your project files, inspect environment variables, and follow your folder structure. Project context — example application

Project: KodeKloud Task Manager (Flask)
Key files Cursor can read: app.py, templates, static assets, requirements.txt, and tests/ folder.
Environment: Python virtual environment, SQLite DB stored in instance/

Below is a short excerpt from app.py to illustrate the kind of file contents the assistant can parse and act on.

# app.py (excerpt)
import csv
import sqlite3
import os
from flask import Flask, render_template, request, redirect, url_for, flash, session, g
from datetime import datetime
import hashlib
import logging

# Initialize Flask app
app = Flask(__name__)
app.config['SECRET_KEY'] = 'dev'  # Change this to a random secret key in production
app.config['DATABASE'] = os.path.join(app.instance_path, 'task_manager.sqlite')

# Ensure the instance folder exists
os.makedirs(app.instance_path, exist_ok=True)

def read_csv(file_path):
    with open(file_path, 'r', newline='') as f:
        csvreader = csv.reader(f)
        for row in csvreader:
            print(row)

This snippet exposes a development SECRET_KEY value. Do not commit secrets to source control — replace with environment-based configuration (e.g., os.environ['SECRET_KEY']) before deployment.

Installing test tooling and updating dependencies

Prompt example: “Install pytest and update requirements.txt” — Cursor will propose the commands and can execute them in the terminal.
Typical generated steps:
1. Activate virtualenv: source venv/bin/activate (macOS/Linux) or venv\Scripts\activate (Windows)
2. Install test libs: pip install pytest
3. Freeze to requirements: pip freeze > requirements.txt

Generating and executing commands interactively

Use the quick command generator (e.g., Ctrl+K) to get a proposed command and then run it in the terminal.
This is “question mode”: ask how to do something, review the generated command, then run it.
Switch suggestion models (e.g., Claude 3.7) to alter suggestion style or thoroughness.

Creating tests — recommended workflow

Ask the assistant to “Create a test for the dashboard” and it may propose creating a tests/ folder and a tests/test_app.py file.
If automated generation fails, create the files manually and paste the assistant’s suggested test content.

Example test for deleting a task:

# tests/test_app.py (example)
def test_delete_task(client):
    # Delete the task
    response = client.post('/task/1/delete', follow_redirects=True)

    assert response.status_code == 200
    assert b'Task deleted successfully!' in response.data
    assert b'Test Task' not in response.data

Running tests — examples and troubleshooting

Be specific with test targets. Running pytest against the wrong path can yield “collected 0 items”.
Wrong example (runs the app module, not the tests):

venvjeremy@MACSTUDIO KodeKloudTaskManager % python -m pytest KodeKloudTaskMan/app.py
============================= test session starts =============================
platform darwin -- Python 3.13.1, pytest-8.3.5, pluggy-1.5.0
rootdir: /Users/jeremy/Projects/KodeKloudTaskManager
collected 0 items

============================== no tests ran in 0.09s
venvjeremy@MACSTUDIO KodeKloudTaskManager %

Correct example — run the tests in the tests/ folder or point to a specific test:

venvjeremy@MACSTUDIO KodeKloudTaskManager % python -m pytest tests/test_app.py
============================= test session starts =============================
platform darwin -- Python 3.13.1, pytest-8.3.5, pluggy-1.5.0
rootdir: /Users/jeremy/Projects/KodeKloudTaskManager
collected 6 items

/Users/jeremy/Projects/KodeKloudTaskManager/KodeKloudTaskMan/app.py:165: DeprecationWarning: The default timestamp converter is deprecated as of Python 3.12; see the sqlite3 documentation for suggested replacement recipes
  tasks = db.execute(query, params).fetchall()

...F.F
= 2 failed, 4 passed, 2 warnings in 0.12s
venvjeremy@MACSTUDIO KodeKloudTaskManager %

To run a single test function:
- python -m pytest tests/test_app.py::test_delete_task

Writing a reusable pytest fixture for the Flask app

Create tests/conftest.py to configure a temporary database and Flask test client:

# tests/conftest.py (example)
import os
import tempfile
import pytest
from KodeKloudTaskMan.app import app, init_db

@pytest.fixture
def client():
    db_fd, db_path = tempfile.mkstemp()
    app.config['DATABASE'] = db_path
    app.config['TESTING'] = True

    with app.test_client() as client:
        with app.app_context():
            init_db()
        yield client

    os.close(db_fd)
    os.unlink(app.config['DATABASE'])

Best practices when using a context-aware terminal

Be specific. Start with a short project overview (framework, structure, main components) and reference concrete file names and paths. Include environment details (Python version, virtual environment, dependencies) and any constraints (libraries to avoid, required versions). This helps the assistant generate accurate commands and tests.

Quick reference — common prompts and commands

Goal	Example prompt / command
Install pytest and update deps	`pip install pytest && pip freeze > requirements.txt`
Run tests in folder	`python -m pytest tests/`
Run single test	`python -m pytest tests/test_app.py::test_delete_task`
Create test file	Ask: “Create tests/test_app.py with a client fixture and CRUD tests”
Debug failing test	Ask: “Run `pytest -k <testname> -q` and show full stacktrace”

More tips

If a generated command fails, refine the prompt with exact filenames, paths, or function names.
Use descriptive prompts like “Run pytest for tests/test_app.py::test_delete_task” rather than generic “run tests”.
When running wide-scope commands (profiling, linting, or large test suites), narrow the scope to reduce noise and speed feedback.

References and further reading

pytest documentation: https://docs.pytest.org/
Flask testing: https://flask.palletsprojects.com/en/latest/testing/
Python tempfile module: https://docs.python.org/3/library/tempfile.html
sqlite3 module: https://docs.python.org/3/library/sqlite3.html

As you iterate with the context-aware terminal, provide more precise context (file paths, versions, constraints). The assistant will return increasingly accurate and runnable suggestions.

Introduction to Cursor

Mastering Autocompletion

Interacting with your Codebase

Inline Editing and Debugging

Terminal Productivity

Understanding and Customizing Cursor

Demo Context Aware Terminal Suggestions

Watch Video