32dots HEIDELBERG AI
Session 3 medium

Give your agents tools and your own LLM

LESSONLesson 3 · ~25 min

🎯Goal. Make your agents more capable by giving them tools and choosing which LLM powers them.

▶ Try this prompt

Give the researcher agent a web-search tool so it can pull current information instead of relying only on the model's memory, and point the crew at the LLM you prefer.

CrewAI is model-agnostic and ships a large library of tools/integrations. You attach tools to agents in code; you pick the model via your .env / config.

  1. 1Add a tool to an agent (for example a search or file-reading tool from CrewAI's tools library). An agent with tools can take actions in the world, not just generate text.
  2. 2Choose your LLM: because CrewAI is model-agnostic, you can run the same crew on different providers or even a local model — you supply the key and model name yourself.
  3. 3Re-run with crewai run and watch the researcher actually call its tool mid-task, then feed the result to the writer.

You'll see. An agent invoking a real tool during its reasoning, and the same crew running on whichever LLM you chose.

💳Cost. Still free at the framework level; tools that hit paid APIs and your LLM usage are billed by those providers. Bringing your own key means you control the spend.

💡Takeaway. Tools turn agents from talkers into doers, and model-agnostic config means you're never locked to one provider.