🔥 Testfire Dev VS Code Extension

🚀 Next-Gen AI Coding with Local Ollama

Testfire Dev is an open-source Visual Studio Code extension that brings advanced AI code completion, chat, and code analysis to your editor—powered by your own local Ollama models (v0.1+). No cloud required. 100% privacy. Fast, free, and developer-friendly.

✨ Features

• AI code completion and chat (local Ollama)
• Works offline—no data leaves your machine
• Supports multiple models (Llama2, etc.)
• Easy setup and configuration
• Open source and MIT licensed

🛠️ Quick Start

1. Install Ollama v0.1+ (Ollama GitHub)
2. Download a model:

ollama pull llama2

3. Start Ollama server:

ollama serve

4. Install this extension in VS Code
5. Configure the extension (set endpoint to http://localhost:11434)
6. Start coding with AI!

For full setup, see OLLAMA_LOCAL_SETUP.md.

📚 Documentation

• Ollama Local Setup Guide
• Extension Features
• Quick Start Routing

🤝 Contributing

We welcome issues, feature requests, and pull requests! See CONTRIBUTING.md (or open an issue to get started).

📢 Keywords

AI, Ollama, VS Code, Code Completion, Llama2, Open Source, Local AI, Privacy, Developer Tools, Chat, Coding Assistant

📄 License

MIT — see LICENSE

⭐️ Star & Share

If you find this project useful, please star the repo and share it with your friends and colleagues!

🔗 Links

• GitHub Repository
• VS Code Marketplace
• Ollama Documentation

TestFire AI Agent 🔥

TestFire.dev - An intelligent AI-powered coding assistant for VS Code with project awareness, multi-file editing, autonomous task execution, and Cursor-like chat interface.

🌟 Features

🎨 Personality & Style

TestFire is designed to be your trusted coding companion with these characteristics:

• Thorough: Examines all relevant files before suggesting changes, analyzing dependencies and patterns across your codebase
• Cautious: Shows you detailed plans and diff previews before making any changes, requiring your approval
• Educational: Explains reasoning behind suggestions and provides learning opportunities with each interaction
• Consistent: Learns and follows your project's existing patterns, coding style, and best practices

🚀 Core Capabilities

💬 Intelligent Chat Interface

• Cursor-style chat panel with streaming responses
• Context-aware conversations that understand your project structure
• Multi-turn conversations with full context retention
• Real-time syntax highlighting in code blocks

📁 Multi-File Operations

• Edit multiple files simultaneously with atomic operations
• Preview all changes with side-by-side diffs
• Accept or reject changes individually or in bulk
• Automatic backup and rollback support
• Undo/redo for all file operations

🤖 Autonomous Agent Mode

• Execute complex, multi-step tasks autonomously
• Automatic planning with step-by-step breakdown
• Progress tracking with live updates
• Plan approval workflow before execution
• Task list visualization

🔍 Deep Project Understanding

• Semantic code search with embeddings
• Dependency graph analysis
• Framework and pattern detection
• Git integration and change tracking
• File tree visualization

📊 Task Tracking & Progress

• Visual task lists with status indicators (⏳ 🔄 ✅ ❌ ⏭️)
• Hierarchical subtasks support
• Real-time progress updates
• Export to markdown checklists
• Automatic cleanup of completed tasks

🎓 Educational Explanations

• Automatic or on-demand explanations for suggestions
• "Why" behind code changes explained clearly
• Pattern and best practice identification
• Learning resources and recommendations
• Confidence levels with reasoning

🔌 Available Tools

This agent can:

• 📁 Read and write files in your workspace
• 🔍 Search across your codebase (text + semantic)
• ✏️ Edit multiple files simultaneously with diff previews
• ▶️ Execute commands in terminal
• 🌐 Access web resources (if needed)
• 📋 Track tasks and todos with visual progress
• 🔄 Analyze dependencies and code relationships
• 🎯 Generate plans before execution
• 💡 Explain reasoning for all suggestions

📦 Requirements

• VS Code: Version 1.109.0 or higher
• AI Provider: One of the following:
  • Ollama (local, recommended): Install Ollama
  • OpenAI API: Requires API key
  • Anthropic API: Requires API key

Recommended Models

• Ollama: deepseek-coder:6.7b, codellama:13b, or qwen2.5-coder:7b
• OpenAI: gpt-4-turbo, gpt-4, or gpt-3.5-turbo
• Anthropic: claude-3-opus, claude-3-sonnet, or claude-3-haiku

🎯 Quick Start

1. Install the extension from VS Code Marketplace
2. Configure your AI provider:
   • Open Settings (Cmd+, or Ctrl+,)
   • Search for "TestFire"
   • Set your provider, model, and API key (if needed)
3. Open the chat: Press Cmd+Shift+I (Mac) or Ctrl+Shift+I (Windows/Linux)
4. Start coding! 🚀

⚙️ Extension Settings

This extension contributes the following settings:

🤖 AI Provider Settings

• testfire.provider: AI provider (ollama, openai, anthropic)
• testfire.providerUrl: Base URL for the AI provider API
• testfire.model: Model to use
• testfire.apiKey: API key (for OpenAI/Anthropic)
• testfire.maxTokens: Maximum tokens in AI response (default: 4096)
• testfire.temperature: Response creativity (0-2, default: 0.7)
• testfire.embeddingModel: Model for semantic search

⚡ Feature Toggles

• testfire.enableHover: AI-powered hover information (default: true)
• testfire.enableCompletions: AI code completions (default: true)
• testfire.enableQuickFix: AI quick fixes and code actions (default: true)

🎨 Behavior Settings

• testfire.showPreviewBeforeApply: Show diff preview and require approval (default: true)
• testfire.showPlanBeforeExecute: Show plan and require approval for autonomous tasks (default: true)
• testfire.enableExplanations: Generate educational explanations (default: true)
• testfire.enableTaskTracking: Show task progress in chat (default: true)
• testfire.autoExplain: Auto-generate explanations for all suggestions (default: false)
• testfire.thoroughnessLevel: Analysis depth (fast, balanced, thorough, default: balanced)

📊 Performance Settings

• testfire.contextMaxChars: Maximum context characters (default: 12000)
• testfire.maxFileSizeKB: Maximum file size to scan in KB (default: 100)

🎮 Usage Examples

💬 Chat Mode

You: "Explain how authentication works in this project"
AI: [Analyzes auth files, explains flow with diagrams]

You: "Add input validation to the login form"
AI: [Shows plan, previews changes, explains why validation is important]

🤖 Agent Mode

You: "Refactor the API layer to use async/await consistently"
AI:
📋 Plan:
  1. ⏳ Analyze current API patterns
  2. ⏳ Identify callback-based code
  3. ⏳ Convert to async/await
  4. ⏳ Update error handling
  5. ⏳ Add tests

[Approve] [Reject]

[After approval...]
  1. ✅ Analyzed 15 API files
  2. 🔄 Converting auth.js...
  3. ⏳ Pending...

🔍 Context-Aware Assistance

[Select code in editor]
You: "Improve this function's performance"
AI: [Analyzes selected code + related files]

    💡 Explanation: This function has O(n²) complexity due to nested loops.

    📊 Preview changes:
    ✅ Added memoization
    ✅ Reduced complexity to O(n)
    ✅ Added benchmark tests

    🎓 Why: Memoization caches results to avoid redundant calculations...

[Approve] [Reject] [Request More Explanation]

⚙️ Configuration Tips

For best results:

✅ Keep related files open in editor for better context
✅ Select relevant code when asking questions
✅ Provide clear, specific task descriptions
✅ Review and approve plans before execution
✅ Give feedback when suggestions aren't perfect
✅ Enable explanations to learn while coding
✅ Use balanced thoroughness for everyday work

The more you use it, the better it understands your coding style and project patterns!

📝 Commands

Access via Command Palette (Cmd+Shift+P / Ctrl+Shift+P):

• TestFire: Open AI Chat - Open the chat interface
• TestFire: Ask AI (Custom Prompt) - Quick AI query
• TestFire: Explain Code - Explain selected code
• TestFire: Improve Code - Get improvement suggestions
• TestFire: Add Unit Tests - Generate tests
• TestFire: Find Bugs & Issues - Analyze for bugs
• TestFire: Refactor Code - Refactor selected code
• TestFire: Run AI Agent - Execute autonomous task
• TestFire: Analyze Project - Full project analysis
• TestFire: Generate Component - Create new component
• TestFire: Re-index Workspace - Refresh project index

🔧 Troubleshooting

Ollama Connection Issues

# Check if Ollama is running
ollama list

# Start Ollama (if not running)
ollama serve

# Pull your model
ollama pull deepseek-coder:6.7b

Slow Responses

• Try a smaller model (e.g., codellama:7b instead of 13b)
• Reduce contextMaxChars in settings
• Set thoroughnessLevel to fast

Out of Memory

• Reduce maxFileSizeKB in settings
• Exclude large files via .testfireignore
• Close unnecessary files in editor

🆕 What's New

v0.1.0 - Initial Release

🎉 New Features:

• ✨ Thorough analysis of project files before suggestions
• 🛡️ Cautious workflow with plan approval and diff previews
• 🎓 Educational explanations for all AI suggestions
• 📋 Task tracking with visual progress indicators
• 🔄 Multi-file editing with atomic operations
• 🤖 Autonomous agent mode for complex tasks
• 💬 Cursor-style chat interface
• 🔍 Semantic code search
• 📊 Dependency graph visualization
• 🌲 File tree integration

🤝 Contributing

Found a bug? Have a feature request? Open an issue!

📄 License

MIT License - See LICENSE file for details

Enjoy coding with TestFire! 🔥