Skip to main content

Artemis Navigator VS Code Extension

Artemis Navigator is the official VS Code extension that brings intelligent planning and strategic AI-driven development directly into your IDE. Unlike generic AI coding tools that guess your intent, Artemis Navigator captures your requirements through targeted questions, generates structured plans, and delivers reliable results without the prompt ping-pong.

Why Artemis Navigator?​

Traditional AI coding assistants often produce quick but unfocused results because they work from vague prompts. Artemis Navigator solves this by:

  • Capturing Intent: Interactive Q&A ensures the AI understands exactly what you need
  • Strategic Planning: Breaking complex features into structured, manageable tasks
  • Context-Aware: Questions are targeted to YOUR codebase, not generic templates
  • Adaptive Plans: Plans evolve with your project, allowing manual adjustments for both human and AI tasks
  • AI-Ready Prompts: Auto-generates highly optimized, repo-aware instructions for any AI coding tool

This is the difference between "vibe coding" and agentic engineering—Artemis transforms AI development from guesswork into a systematic process you can trust.

Installation & Authentication​

Installation and Authentication

After installing the extension, you'll need to authenticate with your Artemis account. The setup process includes:

1. Choose Your Default Task Assignment​

When you first open Artemis Navigator, you can configure how tasks are assigned:

  • My own agent: Bring your own AI assistant (like Claude Code or Cursor). You only pay to generate plans via Artemis Navigator. Code generation is handled by your chosen AI tool.
  • Artemis Cloud Agent: Full-service solution where Artemis generates both plans and code. All-in-one solution with everything handled and billed through Artemis.

You can change this preference later in Settings.

2. Sign In to Your Artemis Account​

Click the Sign in button in the Getting Started panel, or click "Don't have an account yet? Sign Up" to create a new account.

3. Select Your Authentication Method​

Choose how you would like to authenticate:

  • Sign in with Turintech Cloud (Recommended): For most users accessing the production SaaS environment
  • Use your Turintech account: For users with managed cloud environments
  • Sign in with On-Premises: For self-hosted environments. Requires your organization's specific Turintech instance URL (e.g., http://dev.artemis.turintech.ai or https://staging.artemis.turintech.ai)

4. Enter Your Credentials​

Enter your email address and password, then complete the authentication flow.

Once authenticated, you'll have access to all your Artemis projects directly in VS Code.

Getting Started​

1. Access Your Projects​

Once the extension is installed and you've signed in, you'll see the Artemis Navigator icon in the sidebar. Click it to view all your projects.

Projects View

Here you can:

  • Browse all your Artemis projects
  • Import new projects using the + Import button
  • See project creation dates and associated plans
  • Access project repositories and plan history

2. Create a Plan​

Select a project to open it, then use the input field to describe your feature, refactoring task, or structural change.

Create Plan

Step 1: Write your prompt in the text field (e.g., "Add AI assistant that gives personalized advice based on the weather")

Step 2: Click the submit button to start the planning process

This is where Artemis differs from generic AI tools—instead of immediately generating code, it enters Plan mode to understand your requirements.

3. Interactive Requirements Gathering​

Artemis asks intelligent, context-aware questions to capture your intent and requirements.

Requirements Gathering

These questions are:

  • Codebase-specific: Based on YOUR project structure, not generic templates
  • Teaching moments: Revealing feature possibilities and architectural considerations you might not have considered
  • Flexible: Choose from suggested options or type your own custom response

For example, when asked "What kind of personalized advice should the AI assistant provide?", Artemis shows you distinct categories you might not have initially considered—helping you think through the feature more thoroughly.

4. Review Plan Summary​

Before generating the detailed plan, Artemis creates a summary with decision checkpoints.

Plan Summary

This summary includes:

  • Technical Approach: Architectural decisions and implementation strategy
  • Important Notes: Critical considerations like API keys, security, and dependencies
  • Development Flow: Step-by-step execution plan

This is your chance to catch costly mistakes early. It doubles as lightweight documentation, so you can show exactly what was agreed upon.

Review and decide:

  • Click "Yes, proceed" if the plan looks good
  • Click "No, let me adjust" to continue refining requirements

5. View the Plan Dashboard​

Once approved, Artemis generates a comprehensive task breakdown.

Plan Dashboard

The Plan Dashboard shows:

  • Progress bar: Visual tracking of completed vs. remaining tasks
  • Task list: All tasks numbered and organized sequentially
  • Task assignees: Tasks assigned to either User (👤) or Agent (🤖)
  • PR Status: Track which tasks have associated pull requests
  • Actions: "Start Task" buttons for agent tasks, "Open Task" to view details

Notice something important: Not every task is assigned to AI. Artemis automatically decides which tasks are better handled by humans vs. AI based on the nature of the work.

6. Working with Individual Tasks​

Click on any task to expand its details and access AI-optimized prompts.

Task Details

Each task includes:

  • Task Name: Clear, actionable description
  • Assignee: User or Agent assignment
  • Task Description: Detailed requirements and specifications
  • Prompt: AI-ready instructions you can copy
  • Technical Specs: Implementation details specific to your codebase

The secret sauce: These auto-generated prompts are precise, repo-aware instructions designed to give AI agents the clearest, most efficient path to success. No more vague instructions.

7. Dynamic Task Management​

Plans are flexible and can evolve with your project.

Task Management

From the task menu (⋮), you can:

  • Copy Prompt: Get the AI-optimized prompt for any coding tool
  • Edit Task: Modify task details or descriptions
  • Remove Task: Delete tasks that are no longer relevant
  • Add Task above/below: Insert new tasks anywhere in the plan

You can also manually add new tasks at any point—either for yourself or for the Agent. The bottom image shows creating a custom user task with its own description.

8. Execute Tasks with AI Tools​

Copy the optimized prompt and paste it into your preferred AI coding tool (like Claude Code).

Execute with AI

The workflow:

  1. Copy the Artemis-generated prompt
  2. Paste into your AI coding tool (shown: Claude Code interface)
  3. Let the AI execute with precise, context-aware instructions
  4. Review and test the generated code

Because the prompt is highly optimized and includes full technical context, you get much better results than generic prompting.

9. Mark Tasks Complete and Track Progress​

After completing each task, mark it as done to track your progress.

Complete Tasks

Best practices:

  • Test the code before marking complete
  • Commit working code before moving to the next task
  • If using Artemis prompts with other AI tools, mark tasks as "Complete" when done

The image shows 7 of 8 tasks completed, with clean git commits demonstrating incremental, validated progress.

10. View All Completed Tasks​

Switch to the Tasks tab to see your full task history.

Tasks View

This view shows:

  • All tasks with completion status (✓)
  • Task assignees (User 👤 vs. Agent 🤖)
  • "Agent Done" indicators for AI-completed work
  • "Open Task" buttons to review task details and PRs

This provides a complete audit trail of what was built, by whom (human or AI), and when.

11. Final Results​

When all tasks are complete, you'll have a fully implemented feature with clean commits and zero broken builds.

Project Complete

The final view shows:

  • All 8 tasks completed
  • Clean git history with descriptive commit messages
  • Zero broken builds
  • Key Improvements Delivered documentation

Key Improvements Delivered:

  1. Relevance: Context-aware prompts ensure advice matches specific conditions
  2. Quality: Structured format with guidelines improves response consistency
  3. Robustness: 4-layer parsing fallback handles edge cases gracefully
  4. Efficiency: 400 token limit optimizes cost while maintaining quality
  5. Testing: 10 diverse scenarios with documented expected outcomes
  6. Documentation: Complete testing guide for validation and future improvements

This is how AI development should work—strategic thinking first, then precise execution.

Configuration​

Available Commands​

Access these commands via the command palette (Ctrl + Shift + P):

  • Artemis Navigator: Set default Task assignment - Configure automatic task assignment
  • Artemis Navigator: Enable Debug Logging - Enable detailed logs for troubleshooting
  • Artemis Navigator: Open Logs - View extension logs
  • Artemis Navigator: Sign Out - Disconnect from your Artemis account

Tips & Best Practices​

  • Think Before Coding: Use Plan mode for complex features to ensure clarity before implementation
  • Review Decision Checkpoints: The plan summary is your opportunity to catch mistakes early
  • Commit Incrementally: Complete and test each task before moving to the next
  • Use Optimized Prompts: Copy Artemis prompts to use with any AI coding tool for better results
  • Track Progress: Mark tasks complete to maintain visibility into your project status
  • Stay Flexible: Add or remove tasks as your understanding of the feature evolves

Troubleshooting​

Enable Debug Logging​

For troubleshooting issues:

  1. Open the command palette (Ctrl + Shift + P)
  2. Select "Artemis Navigator: Enable Debug Logging"
  3. Reproduce the issue
  4. Access logs via "Artemis Navigator: Open Logs"

Common Issues​

  • Authentication failures: Verify you're using the correct environment URL and credentials
  • Extension conflicts: Uninstall old versions completely before installing the new version
  • Connection issues: Check your network connection and firewall settings

Feedback & Support​

We value your feedback and are committed to continuously improving Artemis Navigator.

  • Rate us on the VS Code Marketplace
  • Feature Requests - Share your ideas for new features
  • Bug Reports - Help us improve by reporting issues

Your input helps us make Artemis Navigator better for everyone!

Last updated on by neda.z