Test

What is Test?

Test is an AI assistant that can use your CLI and Editor to handle complex software development tasks step-by-step. It integrates with VS Code and supports multiple API providers like OpenRouter, Anthropic, OpenAI, Google Gemini, AWS Bedrock, Azure, and GCP Vertex. It is designed for developers who want autonomous coding assistance with human oversight.

How to use Test?

Install the Test extension from the VS Marketplace (VS Code extension). Open the command palette with CMD/CTRL + Shift + P and type "Cline: Open In New Tab" to use it side-by-side with the file explorer. Enter a task and optionally add images; Test then analyzes your project, reads files, and executes commands after granting permission.

Key features of Test

• Create and edit files with diff review and timeline tracking
• Execute terminal commands and monitor output in real-time
• Launch a headless browser for interactive web debugging and testing
• Add custom tools via the Model Context Protocol (MCP)
• Fetch URLs, add problems, files, or folders as context
• Workspace checkpoints to compare and restore snapshots

Use cases of Test

• Convert mockups into functional apps from screenshots
• Fix runtime errors and visual bugs in web applications
• Manage deployments, databases, and tests via terminal commands
• Add custom tools like fetching Jira tickets or managing AWS EC2s

FAQ from Test

What APIs does Test support?

Test supports OpenRouter, Anthropic, OpenAI, Google Gemini, AWS Bedrock, Azure, GCP Vertex, any OpenAI compatible API, and local models through LM Studio/Ollama.

Can Test create its own tools?

Yes, Test can create and install custom MCP servers on demand, e.g., "add a tool that fetches Jira tickets."

How does Test handle file changes?

It creates and edits files directly in the editor with a diff view; all changes are recorded in the file's timeline for easy tracking and reverting.

Does Test require internet access?

It requires an API provider connection (e.g., OpenRouter, Anthropic) to operate, but local models via LM Studio/Ollama can work offline.

How are costs tracked?

The extension keeps track of total tokens and API usage cost for the entire task loop and individual requests.