> For the complete documentation index, see [llms.txt](https://griffin-ai.gitbook.io/knowledge-base/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://griffin-ai.gitbook.io/knowledge-base/ai-agent-framework/griffin-ai-agent-structure.md).

# Griffin AI Agent Structure

The Griffin AI Agent framework is designed to facilitate the creation and execution of AI Agents through a combination of Agent Tasks and Tools. These components are integral to defining an **Agent** (Program), which represents the actual implementation of an agent instance executed within the Griffin AI physical architecture.

<figure><img src="/files/qvLYhFfCThVs2SwrNMik" alt=""><figcaption><p>Figure 4: Structure of Griffin AI Agent Framework</p></figcaption></figure>

**Tools Framework**

At the core of the AI Agent framework is the Tools Framework, a robust set of instruments grouped by domains and accessible to every AI Agent in real-time. These tools enable AI Agents to interact with the external world, including blockchain ecosystems, enhancing their capability to perform a wide range of tasks from simple computations to complex blockchain interactions. &#x20;

&#x20;

**AI Crew**&#x20;

For the execution of more complex tasks, AI Agents can be organized into a multi-agent group known as an AI Crew. Within this configuration, agents collaborate to accomplish a defined set of tasks through directed processes. Each AI Crew is designed to operate using "Processes," a platform entity that manages task execution by coordinating agent activities. This setup allows individual agents within the crew to function cohesively, streamlining their efforts to achieve common objectives efficiently.&#x20;

&#x20;&#x20;

**Process Execution**&#x20;

The execution of tasks within an AI Crew is managed through processes, which can be visualized as a graph containing four essential components:&#x20;

* Entry Point: The initiation phase of the process.&#x20;
* Agents as Nodes: Each agent involved is represented as a node within the process graph.&#x20;
* Orchestration Logic: Directed edges, including conditional edges, represent the orchestration logic guiding the interactions and sequences of tasks among the agents.&#x20;
* End Point: The conclusion or output phase of the process.&#x20;

<figure><img src="/files/UWrcCbd08OeuFZQC4qul" alt=""><figcaption></figcaption></figure>

**Process Execution**&#x20;

Processes within the Griffin AI framework can be implemented in several forms to suit different operational needs, as detailed in the table below.&#x20;

&#x20;

| Name                             | Definition                                                                                                                                                                                                                                                    |
| -------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Sequential**                   | Tasks are executed one after the other, ensuring a systematic completion of tasks.                                                                                                                                                                            |
| **Hierarchical**                 | Tasks are organized in a managerial hierarchy, where delegation and execution follow a structured chain of command. To enable this hierarchical process, a manager Agent  must be specified within the crew, allowing the manager to create and manage tasks. |
| **Consensual Process (Planned)** | This process type aims for collaborative decision-making among agents regarding task execution, introducing a democratic approach to task management within Griffin.AI platform.                                                                              |

Table 9: Types of Process Implementation&#x20;


---

# 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, and the optional `goal` query parameter:

```
GET https://griffin-ai.gitbook.io/knowledge-base/ai-agent-framework/griffin-ai-agent-structure.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.