Link the separate agent traces and grading dataset

Link all completed model runs and final grades

Remove research post link and update code references

Streamline dataset card and link benchmark walkthrough

Clarify 50-task results and link research post

Include all 50 tasks in benchmark results for v1.1

Add API key environment example and setup instructions

Simplify dataset card and shorten intended-use note

Publish FAB v1.0 data room, tasks and dataset card

initial commit