Task gallery
100 tasks across 4 families.
Each task: a realistic brief, a source video, a verifiable rubric. The full task set lives in the GitHub repo.
Assembly
18
tasks
Repair
18
tasks
Sequencing
28
tasks
Repurpose
36
tasks
Walkthrough
What a task actually looks like.
A short walkthrough of one task per family, showing the brief, the source material, and how the verifier scores an agent's output.