Skip to main content

Validate Agent Configuration

Validate the configuration of an agent to ensure it is set up correctly. you can check for any potential issues in the agent's configuration and provides feedback on what needs to be corrected.

Upload YAML Configuration​

You can upload a YAML file containing the agent's configuration. The system will validate the configuration and provide feedback on any issues found.

Python
from svahnar import Svahnar
from pathlib import Path


client = Svahnar()
response = client.agents.validate(
  yaml_file=Path("agent.yaml")
)

print(response)

Validate YAML Configuration String​

Alternatively, you can provide the YAML configuration as a string. This is useful for quick tests or when you don't want to create a separate file.

Python
from svahnar import Svahnar
from pathlib import Path

client = Svahnar()
response = client.agents.validate(
  yaml_string="""
  create_vertical_agent_network:
  agent-1:
    agent_name: information search agent
    agent_function:
      - It is used to search for the information on web or internet and provide the result
    LLM_config:
      params:
        model: gpt-4o-mini
        temperature: 0.5
        max_tokens: 1000
        request_timeout: 600
    tools:
      tool_assigned:
        - name: Tavily
    incoming_edge:
      - Start
    outgoing_edge: []
"""
)
print(response)

Response​

You will receive a clear response indicating:

  1. Whether the configuration is valid
  2. Any issues that need to be addressed

In the Suggestions section, you will find recommendations to resolve any identified issues.

Response
{
  "message": "Validation successful: No issues found!",
  "status": "success",
  "suggestion": None
}

Response Parameters​

  • message: A message indicating the result of the validation process.
  • status: The status of the validation process, indicating whether it was successful or if there were issues.
  • suggestion: Any suggestions or recommendations for improving the configuration. This may include specific changes to make or best practices to follow.