Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
i guess it's up to you to tell the agent what to limit it to. if you/your team are worried about accountability, then don't let it actually commit any code, just have it do everything short of the commit? or depending on your work flow, let it do the commit, but not the PR.. EDIT: also, if an AI commit/PR can kill the prod at 3am, then you really need continuous integration.
reddit Viral AI Reaction 1776613561.0
Coding Result
DimensionValue
Responsibilityuser
Reasoningdeontological
Policynone
Emotionindifference
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_oh3ee80","responsibility":"ai_itself","reasoning":"consequentialist","policy":"regulate","emotion":"fear"}, {"id":"rdc_oh3j1nf","responsibility":"user","reasoning":"deontological","policy":"liability","emotion":"indifference"}, {"id":"rdc_ohf5qu9","responsibility":"distributed","reasoning":"deontological","policy":"regulate","emotion":"fear"}, {"id":"rdc_oi2py6c","responsibility":"company","reasoning":"consequentialist","policy":"industry_self","emotion":"approval"}, {"id":"rdc_oh3e3td","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"indifference"} ]