If you give an agent a task, the typical agentic pattern is that it calls tools in some non-deterministic loop, feeding the tool output back into the LLM, until it deems the task complete. The LLM internalizes an algorithm.

Another way of doing it is the agent just writes an algorithm to perform the task and runs it. In this world, tools are just APIs and the agent has to think through its entire process end to end before it even begins and account for all cases.

Only latter is turing complete, but the former approaches the latter as it improves.

My read was roughly that agents require constraining scaffolding (CLAUDE.md) and careful phrasing (prompt engineering) which together is vaguely like working in a DSL?

If the llm is able to code it, there is enough training data that youight be better off in a different language that removes the boilerplate.

https://en.wikipedia.org/wiki/Clarke's_three_laws