Post

Specs become condensed shortcuts for invoking prompt sequences referencing other files, resources or code. It's like the LLM equivalent of a shell script, but instead of writing scripts to make calls to LLMs, you write prompts which the agent executes, gluing inferences together with code for you

Please(!) do not use this approach for things that affect production systems or require strict correctness. Consider it for things like summarization, code scaffolding, writing initial unit tests and more generative tasks. It could work well when you're already scripting LLM calls.