Post-processing LLM output

Javascript uploaded to a prompt or context file able to scan and modify LLM output, before the user sees it. The use case I have in mind is games that preserve state (i.e. variables) in the output, where the script could conceal the state mechanism from the user = better immersion.