Generate Agents

The luci-cli refined_agent command is used to create or define a refined AI agent within the Luci framework. This example demonstrates how to create an agent focused on building an open-source AI Medical Agent Framework, which can assist in a company specializing in AI medical scribing.

luci-cli refined_agent --name "Research" --objective "Building open source AI Medical Agent Framework how it can help my company focused on ai medical scribe" --precautions "Avoid using outdated information" --task "AI Medical Agent framework experience"

Luci CLI also allows users to run agents directly from a YAML configuration file, which can streamline the execution of tasks by predefining settings.

luci-cli run_agent --yaml-path "agents_configs/Research_research.yaml" --model "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo" --method "call_together"

Explanation of Parameters:

--yaml-path: The path to the YAML configuration file where the agent's parameters are stored.

Example: "agents_configs/Research_research.yaml" points to the configuration file for the Research agent.

--model: Specifies the model to use for the agent.

Example: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo" indicates the use of Meta Llama's AI model.

--method: The method to call for the AI model. In this case, it refers to an asynchronous method.

Example: "call_together" uses the call_together method for interacting with the specified model.