... scarier than my 1,000 GitHub issues. At least my agent failed loudly enough to blow up my inbox, yours basically gaslit the whole team into believing fake metrics. Did you end up adding tests specifically to catch hardcoded stubs, or are you manually reviewing every line of generated scripts now? Comment from Konstantin Tikhaev(@konstantin_tikhaev): @rakhijha  Writing tests for stubs

... Bones 3.1 is out your generators, your templates Your stubs come first. Every make:* command reads the stubs

... moment you'd normally step in. Treating the agent like a chaotic junior dev is the right instinct. The habit that actually holds up is deciding what 'done' means for each sub-feature before generating anything, so a silent stub gets caught against a real spec instead of a green build screen. Catching the quiet failures is the harder meta-skill, more than any prompting framework. Comment from Konstantin Tikhaev(@konstantin_tikhaev): I can deal with broken code. The agent ... ... silently stubs

... offers, while the api will take a request the ui would never send, someone else's id in the params or a field the form doesn't expose. the agent writes the route so the happy path works and stubs the auth check to match. what caught the most for me was hitting my own endpoints with a second account's token and seeing what came back. Comment from Nikolas Dimitroulakis(@nikolas_dimitroulakis): @henry_lindsey  good point! Comment from ...