I run a lot of AI simulations and currently have no way to export the results. When you’re dealing with 100 tests across 10+ metrics — each with detailed descriptions for both passed and failed cases — reading through each test’s results, transcripts, etc. manually becomes unmanageable.
Being able to export test results (transcripts, metrics, pass/fail details) would let me feed them directly into an LLM for analysis, saving significant time and enabling much faster iteration on my AI agents.