Evals / pdf

Compress

3 scenarios, 3 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
Shrink a scan for emailscan.pdf is too big to email. Make a copy called small.pdf under 150KB that is still readable.fail56s · 12.7kpass14s · 8.2kpass13s · 6.1kpass56s · 22.7kpass26s · 14.0knot runnot run
Compress a photo-heavy PDFReduce photos.pdf to less than half its size and save it as light.pdf.pass10s · 3.4kpass12s · 4.5kpass13s · 9.5kpass21s · 11.9kpass16s · 10.5knot runnot run
Compress a folder of PDFsCompress every PDF in the reports folder and put the results in a folder called compressed, keeping the file names.pass15s · 3.9kpass16s · 10.7kpass12s · 10.4kpass26s · 11.6kpass10s · 10.5knot runnot run