Tag4 posts
All posts tagged verification.
My demo refund agent returned 200, claimed SUCCESS and refunded twice. How I catch that: tool twins that record every call, and faults that split the reply from the state.
RealPath's leakage claim is a pytest: delete everything after the anchor, rebuild, diff. What it caught when I broke the cutoff, and two leaks it still misses.
Before the fix, every solar-return prompt my astrology app could build carried an empty chart and still asked for poetry. How I moved the line between the ephemeris and the LLM into code, and where it isn't there yet.
An auto-generated TBMM summary said a member voted on a bill that hadn't been voted on. The journalist held the story for 24 hours. Four days of building a verification layer that can say no.