Discussion: Being able to serialise steps and continue -> Can we use. agent to create it's own tools?

[Maxim-Filimonov]

Inspired by https://jina.ai/news/auto-gpt-unmasked-hype-hard-truths-production-pitfalls/

Looking at the code I believe, it should be possible to implement serialisation. Idea would be simple - dump an algorithm into JSON file.
I do wonder if it's a bigger question on agents being able to abstract their own tasks into concepts and create a list of possible tasks to use.
Like when I was writing a trading bot I kept thinking - why agent cannot create a tool once like "get_technical_indicators" and then reuse it. It keeps writing the tool from scratch every time.
It's like software development before invention of libraries.

[lgrammel]

Developing ways to pause and resume agent runs (and to provide feedback to agents) is something I've been thinking about. As a first step, I want to have a clean step execution tree. That tree will be serializable and is the main component to serializing a full agent run. Such a serialized agent run could be used to resume agent runs. It could also be the foundation for extracting information, maybe even reusable execution fragments as you suggest.