Agent Anatomy
Every autonomous agent has six core components. Click each one to see what it does and why it matters.
Who the agent is. Its name, role, personality, and voice. Identity shapes every decision the agent makes — it's the "soul" of the system.
What the agent knows and remembers. Short-term (current conversation), long-term (stored facts, past decisions), and shared (accessible to other agents).
What the agent can do. API calls, database queries, file operations, sending messages. Tools are the agent's hands — without them, it can only think.
What the agent is trying to achieve. Goals can be persistent (always active) or triggered (activated by events). They drive the decide-act loop.
What the agent must NOT do. Boundaries, safety rules, ethical constraints. Guardrails prevent the agent from going rogue — they're the conscience before the conscience layer.
When the agent runs. Event-driven (reacts to triggers), cron-based (runs on schedule), or always-on (continuously monitoring). Schedule determines autonomy level.
🛠️ Build Your First Agent
This JSON structure is how agent configurations are typically stored. Each field maps to one of the six components above. JSON (JavaScript Object Notation) uses curly braces for objects, square brackets for lists, and quotes for text values.
Match Component to Description
Tap one on the left, then its match on the right