Re:Anthropic Claude 3.5 can create icalendar files, so I did...
by dbaupp on Sunday August 25, 2024 @05:57AM
(#41346027)
In addition to what you say, it can also be easier for a (appropriately-skilled) human to verify a small program than to verify voluminous parsing output, plus, as you say, there's the semi-automated "verification" of a very-wrong program failing to execute.
Also, the LLM is more likely to fail spectacularly by hallucinating APIs when writing a script, and more likely to fail subtly on parsing tasks.