Tempools logoTempools Docs

Cline

Use Tempools from Cline

Cline is an AI coding agent that lives in your editor and terminal — it reads and writes files, runs commands, and drives a browser. It supports VS Code, Cursor, Windsurf, JetBrains, Antigravity, and a standalone CLI. Tempools plugs in as either an OpenAI Compatible or Anthropic provider.

Prerequisites

  • A supported editor with the Cline extension installed
  • A Tempools API key from the console

Configuration

Open settings

Click the Cline icon in the sidebar, then the gear in the top-right of the panel to open Settings.

Pick the provider

Either of the two paths below works — pick one:

  • API Provider: OpenAI Compatible
  • Base URL: https://api.tempools.com
  • API Key: paste your Tempools key (starts with sk-tb-)
  • Model: enter the exact card title from the Models page or the Console → Models catalog

If Tempools hasn't registered metadata for your chosen model, expand Model Configuration to set context window, max output tokens, Vision / Computer Use, and pricing.

  • API Provider: Anthropic
  • Check (Optional) Custom Base URL and enter https://api.tempools.com
  • API Key: paste your Tempools key
  • Model: pick a Claude-family model from the dropdown

This path only reaches Claude models. For mixed-provider routing, use OpenAI Compatible.

(Optional) Split Plan and Act models

In Settings, enable Use different models for Plan and Act. Plan mode (read-only, for strategy) and Act mode (execution, file edits and commands) can then use different models. In practice, Plan works best with a reasoning-oriented model and Act with a faster coding-oriented one; use the exact card titles from the models catalog.

Verify

Open Cline's panel and try a simple task, e.g. "List the files in this directory and summarize the project". A coherent reply means you're set.

Cline can run shell commands and edit files on its own. Test in a scratch directory first, and audit the scope carefully before enabling Auto-approve. MCP, Memory Bank, and Checkpoints keep writing into context — watch your token spend on long tasks.

Stuck? See the FAQ.

On this page