> For the complete documentation index, see [llms.txt](https://whitepaper.aitech.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://whitepaper.aitech.io/agentforge/tools/thinking.md).

# thinking

The Thinking tool encourages AI models to engage in explicit reasoning before responding to complex queries. By providing a dedicated space for step-by-step analysis, this tool helps models break down problems, consider multiple perspectives, and arrive at more thoughtful conclusions.

Research has shown that prompting language models to "think step by step" can significantly improve their reasoning capabilities. According to [Anthropic's research on Claude's Think tool](https://www.anthropic.com/engineering/claude-think-tool), when models are given space to work through their reasoning explicitly, they demonstrate:

* **Improved problem-solving**: Breaking complex problems into manageable steps
* **Enhanced accuracy**: Reducing errors by carefully working through each component of a problem
* **Greater transparency**: Making the model's reasoning process visible and auditable
* **More nuanced responses**: Considering multiple angles before arriving at conclusions

In Agent Forge, the Thinking tool creates a structured opportunity for your agents to engage in this kind of deliberate reasoning. By incorporating thinking steps into your workflows, you can help your agents tackle complex tasks more effectively, avoid common reasoning pitfalls, and produce higher-quality outputs. This is particularly valuable for tasks involving multi-step reasoning, complex decision-making, or situations where accuracy is critical.

## Usage Instructions

Adds a step where the model explicitly outlines its thought process before proceeding. This can improve reasoning quality by encouraging step-by-step analysis.

## Tools

### `thinking_tool`

Processes a provided thought/instruction, making it available for subsequent steps.

#### Input

| Parameter | Type   | Required | Description                                                                         |
| --------- | ------ | -------- | ----------------------------------------------------------------------------------- |
| `thought` | string | Yes      | The thought process or instruction provided by the user in the Thinking Step block. |

#### Output

| Parameter             | Type   | Description                  |
| --------------------- | ------ | ---------------------------- |
| `acknowledgedThought` | string | Acknowledged thought process |

## Notes

* Category: `tools`
* Type: `thinking`


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://whitepaper.aitech.io/agentforge/tools/thinking.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.