## Create

`agents.create(AgentCreateParams**kwargs)  -> AgentCreateResponse`

**post** `/v2/gen-ai/agents`

To create a new agent, send a POST request to `/v2/gen-ai/agents`. The response body contains a JSON object with the newly created agent object.

### Parameters

- **anthropic\_key\_uuid:** `str`

  Optional Anthropic API key ID to use with Anthropic models

- **description:** `str`

  A text description of the agent, not used in inference

- **instruction:** `str`

  Agent instruction. Instructions help your agent to perform its job effectively. See [Write Effective Agent Instructions](https://docs.digitalocean.com/products/genai-platform/concepts/best-practices/#agent-instructions) for best practices.

- **knowledge\_base\_uuid:** `List[str]`

  Ids of the knowledge base(s) to attach to the agent

- **model\_uuid:** `str`

  Identifier for the foundation model.

- **name:** `str`

  Agent name

- **openai\_key\_uuid:** `str`

  Optional OpenAI API key ID to use with OpenAI models

- **project\_id:** `str`

  The id of the DigitalOcean project this agent will belong to

- **region:** `str`

  The DigitalOcean region to deploy your agent in

- **tags:** `List[str]`

  Agent tag to organize related resources

### Returns

- `class AgentCreateResponse`

  Information about a newly created Agent

  - **agent:** `Optional[APIAgent]`

    An Agent

### Example

```python
from do_gradientai import GradientAI

client = GradientAI()
agent = client.agents.create()
print(agent.agent)
```