Tools

Tools allow Continue to take action in your IDE and beyond (when you give permission). Currently, they are only supported for the following providers:

• Anthropic - see recommended models here.
• Ollama - see recommended models here.
• OpenAI
• Gemini

To use tools, click on the icon in the input toolbar like below.

To let you balance speed and safety, each tool can be set to 1 of 3 modes:

• Automatic: When the LLM requests to use the tool, Continue will automatically call it and send the response to the LLM.
• Allowed: When the LLM requests to use the tool, Continue will first give you the opportunity to "Cancel" or "Continue" with the tool.
• Disabled: The LLM will not know about or be able to use the tool.

Custom tools

Currently custom tools can be configured using the Model Context Protocol, a standard proposed by Anthropic to unify prompts, context, and tool use.

MCP Servers can be added to hub Assistants using mcpServers blocks. You can explore available MCP server blocks here.

To set up your own MCP server, read the MCP quickstart and then create an mcpServers block or add the following to your config file:

YAML

config.yaml

mcpServers:
  - name: My MCP Server
    command: uvx
    args:
      - mcp-server-sqlite
      - --db-path
      - /Users/NAME/test.db

JSON

config.json

{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "transport": {
          "type": "stdio",
          "command": "uvx",
          "args": ["mcp-server-sqlite", "--db-path", "/Users/NAME/test.db"]
        }  
      }
    ]
  }
}