Loading patterns…
Structured Reflection (Think Tool)
A dedicated no-op "think" tool (Anthropic) declared in the agent tool schema that the model calls mid-trajectory to append structured reasoning between a tool result and the next action. The call has no external effect and returns nothing useful: it simply gives the model a checkpoint to re-read policy, verify constraints, and plan before acting inside a long tool-call chain. Anthropic reports a 54% relative improvement on the tau-bench airline pass^1 metric (0.570 versus a 0.370 baseline) when the tool is paired with a prompt that shows how to use it. Distinct from `metacognitive-monitoring`: that entry is the general capability of monitoring one's own reasoning, whereas the think tool is a concrete tool-schema mechanism for it, separate from extended or inference-time thinking that happens before the first response.
In 30 seconds
- What
- Declares a no-op tool in the schema that the model calls mid-trajectory to pause, re-read policy, and plan before the next action.
- When to use
- Long tool-call chains where the model must verify constraints or policy compliance before committing to an irreversible action.
- Watch out
- The model may skip calling think if the prompt example is weak or absent; effectiveness depends entirely on showing the model when and how to use it.
Ask the AI expert about this pattern
Opens the assistant with your question prefilled. You review it before sending.
Structured Reflection (Think Tool): Overview
A dedicated no-op "think" tool (Anthropic) declared in the agent tool schema that the model calls mid-trajectory to append structured reasoning between a tool result and the next action. The call has no external effect and returns nothing useful: it simply gives the model a checkpoint to re-read policy, verify constraints, and plan before acting inside a long tool-call chain. Anthropic reports a 54% relative improvement on the tau-bench airline pass^1 metric (0.570 versus a 0.370 baseline) when the tool is paired with a prompt that shows how to use it. Distinct from `metacognitive-monitoring`: that entry is the general capability of monitoring one's own reasoning, whereas the think tool is a concrete tool-schema mechanism for it, separate from extended or inference-time thinking that happens before the first response.
- No-op tool in the schema: no side effect and no new data returned
- Called mid-trajectory, between a tool result and the next action
- Creates explicit reflection checkpoints inside long tool-call chains
- Used to re-check policy compliance and validate plans before acting
- Cheap and low-code: one tool definition plus a usage-guiding prompt
- Reported 54% relative pass^1 gain on the tau-bench airline domain
The Agent Architect
One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.
Weekly email, one-click unsubscribe. We only use your address to send the briefing.
References
The papers, specifications, and repositories this pattern is based on.
From the engineer behind this catalog
Get your agent architecture reviewed
This page documents one pattern. Your system runs dozens, and most failures live in how they fit together. Have the whole design reviewed against the 288 patterns in this catalog: architecture, reliability, evaluation and cost, every finding mapped to the pattern that fixes it.
€750 instead of €1,500, one week, written report and walkthrough call, until 30 September