3 posts tagged with "configuration"

An AI agent that can read files, run shell commands, and hit external APIs is useful. An AI agent that does all of that when you only asked it to answer a question is a problem.

Modes are how Toolpack SDK draws that line. A mode defines three things: which tools the AI can access, what system prompt shapes its behavior, and whether the workflow engine (planning, step execution, retries) is active. You set the mode once at init or switch it at runtime, and the SDK enforces the constraints on every request.

Toolpack SDK supports four built-in AI providers out of the box: OpenAI, Anthropic, Gemini, and Ollama. You use your own API keys directly — no subscriptions, no middleman. This post covers everything you need to know about configuring providers, switching between them at runtime, and listing available models.

Configuration is the backbone of any flexible SDK. Toolpack SDK implements a sophisticated hierarchical configuration system that balances simplicity for quick starts with power for complex production deployments. Whether you're integrating AI into your app or managing multiple projects, understanding how configuration works will unlock the full potential of the SDK.

Let's dive deep into how toolpack.config.json works, the three-tier hierarchy, and best practices for managing configurations across different environments.