Skip to content

Think

Give the agent a scratchpad to reason through a problem step by step.

Provides a “think” tool the agent can call with its reasoning as input; the reasoning text is simply returned unchanged. It is a simple, read-only scratchpad tool with no side effects and no approval gate.

Connect this to Tools Agent nodes when you want the agent to explicitly reason through a problem step by step before acting, which can improve reliability on multi-step tasks.

This node has no editable settings.

Returns a tool dependency named think that agent nodes can call. Calling it with a thought string returns that same string unchanged, giving the agent a scratchpad for step-by-step reasoning.

  • No explicit credential or node dependency is declared in the node description.

Connect Think to a Tools Agent’s tools input alongside other tools. During complex or multi-step tasks, the agent can call the think tool to lay out its reasoning before deciding which tool to call next.

  • Check that required settings are present before running the node.
  • If the node uses browser page data, run it on the target tab after the page has loaded.
  • If it calls an external service, verify credentials, permissions, and rate limits.
  • No dedicated source test was found next to this node; verify behavior manually when changing this page.