The "think" tool: Enabling Claude to stop and think in complex tool use situations
Importance: 3 | # | ai, claude
As we continue to enhance Claude's complex problem-solving abilities, we've discovered a particularly effective approach: a "think" tool that creates dedicated space for structured thinking during complex tasks.
Introduce a dummy tool called "think", and let a non-reasoning model use it for certain complex tasks, and that tends to work surprisingly well on a specific type of problems.
When to use the "think" tool
Based on these evaluation results, we've identified specific scenarios where Claude benefits most from the "think" tool:
- Tool output analysis. When Claude needs to carefully process the output of previous tool calls before acting and might need to backtrack in its approach;
- Policy-heavy environments. When Claude needs to follow detailed guidelines and verify compliance; and
- Sequential decision making. When each action builds on previous ones and mistakes are costly (often found in multi-step domains).
When not to use the "think" tool
Whereas the “think” tool can offer substantial improvements, it is not applicable to all tool use use cases, and does come at the cost of increased prompt length and output tokens. Specifically, we have found the “think” tool does not offer any improvements in the following use cases:
- Non-sequential tool calls. If Claude only needs to make a single tool call or multiple parallel calls to complete a task, there is unlikely to be any improvements from adding in “think.”
- Simple instruction following. When there are not many constraints to which Claude needs to adhere, and its default behaviour is good enough, there are unlikely to be gains from additional “think”-ing.
Simon Willison describes it as a 'no-op'.