Evals / video

Cut and join

4 scenarios, 4 run by an agent. Each run shows its result, time and new tokens.

ScenarioCodexwithout deftCodexwith deft, skill installedCodexwith deft, skill as agents.mdClaude Opus 5.5without deftClaude Opus 5.5with deft, skill installedClaude Codewithout deftClaude Codewith deft
Cut a clipCut seconds 2 to 5 out of bunny.mp4 and save them as clip.mp4.pass11s · 5.1kpass14s · 9.4kpass13s · 5.8kpass10s · 8.6kpass16s · 10.4knot runnot run
Join two clipsJoin bunny.mp4 and jellyfish.mp4, in that order, into both.mp4.pass18s · 13.9kpass16s · 9.5kpass11s · 5.2kpass21s · 9.4kpass17s · 9.8knot runnot run
Join with a crossfadeJoin bunny.mp4 and jellyfish.mp4 with a one second fade between them into joined.mp4.pass21s · 10.3kpass19s · 28.4knot runpass20s · 9.4kpass26s · 10.3knot runnot run
Cut into equal partsCut bunny.mp4 into 3 second parts in a folder called parts.fail17s · 6.3kpass18s · 15.9knot runpass33s · 11.2kpass12s · 10.9knot runnot run