AI Assist for Custom Code Generation
Orchestrate introduces AI Assist empowers you to leverage generative AI and generate Python code using prompts. You can use the AI Assist-generated Python code to develop custom action and condition nodes to build playbooks. Using AI Assist, you can develop custom code with speed and accuracy that significantly reduces manual coding time and improves productivity.
You can use AI Assist to generate Python code if the feature is enabled in Administration > Configuration. For more information, see Configure General Settings.
How does it work?
Analysts often need to write custom Python code to parse important alerts, chunk lists, convert timestamps, and many other cases when creating a playbook.
For example, you want to write a function to parse IP address details received from an SIEM JSON alert. You can use the following prompt in the AI Assist tool to generate a custom Python code. The following graphic shows an example of custom code generation by AI Assist to parse source IP addresses from an SIEM alert.
We have described a few examples of AI Assist prompts and the responses to help you get started. For more information, see Example Prompts for AI Assist.
Where can you use AI Assist?
You can create custom Python code on custom action nodes and custom condition nodes.
To know how to use AI Assist for custom action nodes, see Action Nodes.
To know how to use AI Assist for custom condition nodes, see Condition Node.
Daily Token Limit for AI Assist
The AI Assist feature has a daily limit on the number of tokens that can be consumed from every Orchestrate tenant. If the daily token consumption limit for AI Assist is reached, the backend will generate an error message stating that you have reached your daily token limit for AI Assist. The token limit will reset at midnight (00:00 Hrs, UTC) every day automatically.