Drawbly

Drawbly · Your ideas, in your own lines

AI agent workflow

See the loop between a model and the tools its application allows.

AI agent workflow: a user request reaches an agent runner, which asks a model what to do, checks an optional tool call, returns the tool result, and produces an answer.
Use this starter to explain one tool-assisted task, rather than drawing a vague cloud of agents.

What this diagram shows

An application runs the agent. It gives the model a goal, instructions and available tools. The model can respond directly or request a tool call. The application checks and executes an allowed call, returns the result to the model and repeats until the model finishes or a limit is reached. An answer should identify which tool result supports its claim. This is a conceptual flow, not a particular provider API.

Follow a concrete tool call in How an AI agent uses a tool.

Make it fit your system

  1. Replace “Check a trace” with one real tool your application can call.
  2. Label what permission or argument check happens before the call.
  3. Add one failure or stop condition that matters for your task.

Click Use this template to open the drawing. Choose Text to change a label, or Select to move a mark. If a drawing is already open, you can keep it or open the template with a recoverable copy.

Watch the explanation unfold

A captioned walkthrough, without narration. Open Video on the canvas to change the replay pacing or record your own explanation.

Technical reference: Anthropic: Building effective agents.

The drawing and supplied marks are editable example content. No account or cloud saving is required. Free exports include Drawbly credit.

← All templates · Render a recipe locally