Agents Overview

An agent is a configured assistant that plans and executes tasks toward a goal. Unlike a workflow builder, an agent decides how to reach an outcome at runtime — it is not a fixed sequence of steps.

The Execution Loop

When an agent receives a task, the dynamic planning engine runs a Plan → Execute → Reflect loop:

  1. Plan — decompose the goal into steps and decide what each needs
  2. Execute — call tools, query knowledge bases, and run skills inside the sandbox
  3. Reflect — evaluate results, branch on conditions, and re-plan if needed

The loop continues until the goal is met or the agent reports it cannot proceed.

What Makes Up an Agent

FieldDescription
nameDisplay name of the agent.
instructionsSystem prompt describing the agent's role, constraints, and the outcomes it should produce.
execution_modeHow aggressively the agent plans and spends compute (e.g. balanced).
modelsThe model(s) backing planning and execution.
knowledge_basesKnowledge bases the agent can retrieve from.
skillsPackaged capabilities the agent can recall.
tool_categoriesWhich categories of tools the agent may use.
suggested_promptsStarter prompts shown to users.

Instant vs. Planned Execution

For simple use cases — chat assistants or embedded AI features — an agent can run tool-enabled LLM calls instantly with no planning overhead. For complex goals, the full planning engine decomposes and orchestrates multiple steps. Start simple and scale up when needed.

Publishing & Sharing

Agents can be published to make them available to others, embedded as a chat widget on allowed domains, or driven through the Workspace API with a runtime key.

Next Steps