PRO COURSE · 24 MIN · EXPERT

Building Simple Agents (ReAct)

Thought → Action → Observation — the foundation of every modern agent, with halt conditions that don't burn $500.

WHAT YOU'LL BE ABLE TO DO

• Write tool descriptions that get selected reliably (vs. the 60%-selection default)
• Design halt conditions that stop runaway loops BEFORE the credit-card alert
• Apply Schon's reflection-in-action: each Thought MUST evaluate the prior Observation
• Know when to NOT use an agent loop (~30% of 'agent' use cases are over-engineered)
• Walk away with a 7-node ReAct canvas you can drop on any research / extraction agent

HOW THE COURSE RUNS

1. Mindset
2. Choose between two prompts
3. See the transformation
4. Practise in a playground
5. Pass 3 gated questions
6. Take the template to the Studio