Installation
First, install the lyzr-agent-api package:
pip install lyzr-agent-api
For more detailed installation instructions, refer to the installation guide on PyPI.
Initializing the Client
To start using the API, initialize the AgentAPI client with your Lyzr API key:
from lyzr_agent_api.client import AgentAPI
client = AgentAPI(x_api_key="your-lyzr-api-key")
Environment
Create an Environment
The environment is a fundamental building block of the Lyzr Agent API. It defines the modules, features, tools available, and other configurations for your agent.For more,click here.
from lyzr_agent_api.models.environment import EnvironmentConfig, FeatureConfig
environment_config = EnvironmentConfig(
name="Test Environment",
features=[
FeatureConfig(
type="SHORT_TERM_MEMORY",
config={},
priority=0,
)
],
tools=[],
llm_config={
"provider": "openai",
"model": "gpt-4o-mini",
"config": {
"temperature": 0.5,
"top_p": 0.9,
},
"env": {
"OPENAI_API_KEY": "your-openai-api-key"
}
},
)
environment = client.create_environment_endpoint(json_body=environment_config)
print(environment)
# The response will include the environment ID, e.g. {'environment_id': '6wjbwhekndjxxxxx'}
Update an Environment
This function updates the configuration of an existing environment by passing in the environment ID (env_id) and a JSON body containing the new environment settings (json_body).
response = client.update_environment_endpoint(
env_id="env-id",
json_body=environment_config,
)
Get an Environment
This function retrieves a list of all available environments. It returns the details of each environment, allowing you to view their configurations and statuses.
client.get_environments_endpoint()
Agent
Create an Agent
Once the environment is set up, create an agent within that environment. The agent uses the environment ID to operate and can be customized with a system prompt and a description:
from lyzr_agent_api.models.agents import AgentConfig
agent_config = AgentConfig(
env_id="environment-id", # Replace with the actual environment ID
system_prompt="This is a system prompt.",
name="Test Agent",
agent_description="Description of the test agent",
)
agent = client.create_agent_endpoint(json_body=agent_config)
print(agent)
# The response will include the agent ID, e.g. {'agent_id': '66fcghvhxxxxxx'}
Update an Agent
This function updates the configuration of an existing agent by providing the agent ID (agent_id) and a JSON body containing the new settings (json_body). It allows you to modify an agent’s behavior or properties dynamically.
response = client.update_agent_endpoint(
agent_id=agent_id,
json_body=agent_config
)
Get an Agent
This function retrieves a list of all available agents. It returns the details and configurations of each agent, allowing you to review and manage them.
client.get_agents_endpoint()
Delete an Agent
This function deletes an agent by specifying its agent_id.
response = client.delete_agent_by_id(
agent_id=agent_id # Example agent ID to delete
)
Inference
Chat with the Agent
After creating the agent, you can initiate a chat session with it. Provide the user ID, agent ID, and a message to start the conversation:
from lyzr_agent_api.models.chat import ChatRequest
response = client.chat_with_agent(
json_body=ChatRequest(
user_id="user-id",
agent_id="agent-id", # Replace with the actual agent ID
message="Hello",
session_id="session-id",
)
)
print(response)
Execute Task with Agent
from lyzr_agent_api.models.chat import ChatRequest
response = client.create_chat_task(
json_body=ChatRequest(
user_id="your-user-id",
agent_id="your-agent-id",
message="scrap the content of https://www.lyzr.ai/ and give summary of it.",
session_id="your-session-id",
)
)
print(response) #e.g {'task_id':'hsjshxxxxxxxx'}
Get Task Status and Response
response_status = client.get_task_status(
task_id=response.task_id, # Use the task_id from the create response
)
print(response_status)
