Hi all,
Thank you for this great package!
This is s neophyte question, but I figure I should ask anyway !
One interesting and usefull feature of the openai agent sdk is the concept of RunContext and toolContext objects
https://github.com/openai/openai-agents-python/blob/main/docs/context.md.
They basically provide objects that are accessible to all tool calls (or hooks) within a run of the model. This provides a way to get some private (likely complex objects) data accessible to tools calls, link to specific runs but not accessible to the model. This is also a way to have side effects (like logging)
In the same vein, beside wrapping the tools calls ourselves with logging and other facilities, are they ways to have register hooks https://openai.github.io/openai-agents-js/openai/agents/classes/agenthooks/?
Thank you in advance
Hi all,
Thank you for this great package!
This is s neophyte question, but I figure I should ask anyway !
One interesting and usefull feature of the openai agent sdk is the concept of RunContext and toolContext objects
https://github.com/openai/openai-agents-python/blob/main/docs/context.md.
They basically provide objects that are accessible to all tool calls (or hooks) within a run of the model. This provides a way to get some private (likely complex objects) data accessible to tools calls, link to specific runs but not accessible to the model. This is also a way to have side effects (like logging)
In the same vein, beside wrapping the tools calls ourselves with logging and other facilities, are they ways to have register hooks https://openai.github.io/openai-agents-js/openai/agents/classes/agenthooks/?
Thank you in advance